Salesforce Unveils Agentforce 360: Ushering in the &#x27;Agentic Enterprise&#x27; Era

Salesforce CEO says its AI program Agentforce is "part and parcel" of the company

In a Tuesday interview with CNBC's Jim Cramer, Salesforce CEO Marc Benioff said his company's artificial intelligence program, Agentforce, is central to business. "People don't understand that Agentforce is part and parcel of Salesforce," Benioff said, speaking to Cramer from Salesforce's annual conference, Dreamforce, in San Francisco. "It is the core of every product we make now, it is the platform." Benioff suggested that the Agentforce platform has developed significantly since the company first introduced it a little over a year ago. The software helps companies use autonomous bots to perform tasks and is meant to increase efficiency. Benioff said Agentforce has "10s of 1000s of deployments, and those customers are all ready to get to another level of execution." Salesforce stock has suffered this year as investors look for faster revenue growth. Shares are currently down more than 28% year-to-date. Cramer asked Benioff about concerns that "AI is eating software," meaning that AI is negative for enterprise software outfits because companies could use generative AI to build their own software tools. Benioff pushed back against such a narrative, and he also named a number of major Salesforce customers including Dell, Williams-Sonoma and PepsiCo. "Now's the time that enterprise and AI are coming together to transform, to become new agentic enterprises," Benioff said.

[4]

Salesforce's Agentforce software is coming to OpenAI's ChatGPT

Salesforce CEO Marc Benioff participates in an interview during the World Economic Forum in Davos, Switzerland, on Jan. 22, 2025. Salesforce is ramping up partnerships with leaders in generative artificial intelligence as investors continue to fear that the software company faces business risks due to the rapid growth of AI. Just ahead of its annual Dreamforce conference in San Francisco, Salesforce said Tuesday it will enable the use of AI models from OpenAI and Anthropic inside its Agentforce 360 software. A day earlier, Salesforce expanded Agentforce beyond text chats to also handle voice calls. "The way people are going to interact with software is going to fundamentally shift," said Brian Landsman, CEO of Salesforce's AppExchange business and executive vice president of partnerships, in an interview. The interaction could be in ChatGPT or in Slack, he said. Salesforce will collaborate with Anthropic to bring Agentforce 360 into Claude, Landsman added. Shares of Salesforce are down about 26% this year, while the S&P 500 index has gained 13%, as Wall Street seeks faster revenue growth from the cloud software company. So far, Agentforce revenue has been "modest," Morgan Stanley analysts, who have the equivalent of a buy rating on Salesforce, wrote in a Monday note. Large software companies are increasingly turning to popular AI model developers for new capabilities. Atlassian, Datadog and Intuit have previously signed deals with OpenAI, and Microsoft has invested almost $14 billion in the company. In September, Databricks committed to spending $100 million on OpenAI models. As part of Salesforce's announcement, customers will be able to access corporate information in Agentforce 360 and create charts in Tableau through the ChatGPT assistant, which has more than 800 million weekly users. Last week OpenAI announced a software development kit for integrating third-party applications into ChatGPT. Companies working with both OpenAI and Salesforce will be able to sell products through ChatGPT's instant checkout feature, the press release said, adding that more detail will be provided in the coming months. Salesforce corrected a press release from earlier in the day, when it said Agentforce 360 apps and Agentforce Commerce in ChatGPT "will be live later this year." Salesforce plans to work with Anthropic on selling products for regulated industries, starting with financial services. OpenAI said last month that ChatGPT users would be able to purchase products from U.S. Etsy sellers and Shopify merchants. Meanwhile, Salesforce said its engineering organization is adopting Anthropic's Claude Code programming product. "We plan to continue to go much deeper with these partners over time," Landsman said. Salesforce CEO Marc Benioff has been defending his company's position in the AI boom. And on last month's earnings call, he said Anthropic and OpenAI both use Salesforce tools. "All these next-generation AI companies ranging from OpenAI to Anthropic to everyone are on Slack," Benioff, who is also Salesforce's co-founder, told analysts. "And it is incredible how they've used that as their operating system and as their platform to run their companies."

[5]

TechRadar

"This is the next revolution" - Salesforce CEO Marc Benioff hails the arrival of the 'agentic enterprise'

Salesforce unveils more on Agentforce 360 and other new offerings Salesforce CEO Marc Benioff has declared the age of the "agentic enterprise", with AI playing a larger role than ever before for firms of all sizes. The company co-founder and CEO opened Dreamforce 2025 with a typically rambunctious keynote as he talked through its latest round of updates and releases - top of which was the new Agentforce 360 platform. But top of mind, of course, was the growing role of AI tools for enterprises, as human workers and agents will have to work together in this mix - which Salesforce calls the 'Agentic Enterprise'. Welcoming attendees to, "the biggest, most exciting Dreamforce ever", Benioff was in a typically charismatic mood, racing through the huge amounts of new additions to Salesforce's platforms at a rapid pace. Benioff noted there is a "very real" divide in using agentic AI, adding that simply deploying enterprise AI models isn't enough, if solutions aren't integrated and grounded in proper governance. "This agentic divide is split - it's a bifurcation. Your customers are elevated, now what about your enterprise?" "We want to connect with our customers in a brand new way...the rate and growth of innovation has been awesome," Benioff noted. However he highlighted that when at home on his phone, "using amazing AI and LLMs", the experience is "so much worse" when he gets to the office and logs on to AI tools there. Salesforce wants to play a key role in advancing AI technology, particularly when it comes to Agentforce, Benioff added, saying, "What we realized at Salesforce is we have to be customer zero...we have to be able to go from one, to two, to three, to four, to five and show you what we're doing." "We do 11 trillion emails a year at Salesforce, but each of those emails is a one way conversation - but what if each of those emails was a two-way conversation?" "AI alone is not enough. It's not enough just to have an LLM. It needs (Agentforce) capabilities to connect it, to give it the context, to give it the guardrails, and all of the critical pieces." "You have got to get your data right. You have got to get to more integrated solutions. You have got to get the priorities right. You have to get the governance right." "We all saw that just making our own models or rolling it ourselves, or DIYing it, isn't going to do it. We need to be doing our work at Salesforce for you, and we need to all be on this journey together to get to this Better Place. That in our vision, is the Agentic Enterprise." "This is the next revolution," Benioff declared, "we've gone through predictive AI, and we're now entering this new agentic AI revolution." Benioff highlighted how the company has continued its 1% pledge, where it donates 1% equity, 1% of its time, and 1% of its product to good causes, including a new $30 million committed to education and AI literacy - meaning Salesforce has now given away $841 million, committed to 10 million nonprofit hours, $300 million in reskilling training and $150 million for local schools in San Francisco and Oakland.

[6]

VentureBeat

Salesforce bets on AI 'agents' to fix what it calls a $7 billion problem in enterprise software

As 50,000 attendees descend on Salesforce's Dreamforce conference this week, the enterprise software giant is making its most aggressive bet yet on artificial intelligence agents, positioning itself as the antidote to what it calls an industry-wide "pilot purgatory" where 95% of enterprise AI projects never reach production. The company on Monday launched Agentforce 360, a sweeping reimagination of its entire product portfolio designed to transform businesses into what it calls "agentic enterprises" -- organizations where AI agents work alongside humans to handle up to 40% of work across sales, service, marketing, and operations. "We are truly in the agentic AI era, and I think it's probably the biggest revolution, the biggest transition in technology I've ever experienced in my career," said Parker Harris, Salesforce's co-founder and chief technology officer, during a recent press briefing. "In the future, 40% of the work in the Fortune 1000 is probably going to be done by AI, and it's going to be humans and AI actually working together." The announcement comes at a pivotal moment for Salesforce, which has deployed more than 12,000 AI agent implementations over the past year while building what Harris called a "$7 billion business" around its AI platform. Yet the launch also arrives amid unusual turbulence, as CEO Marc Benioff faces fierce backlash for recent comments supporting President Trump and suggesting National Guard troops should patrol San Francisco streets. Why 95% of enterprise AI projects never launch The stakes are enormous. While companies have rushed to experiment with AI following ChatGPT's emergence two years ago, most enterprise deployments have stalled before reaching production, according to recent MIT research that Salesforce executives cited extensively. "Customers have invested a lot in AI, but they're not getting the value," said Srini Tallapragada, Salesforce's president and chief engineering and customer success officer. "95% of enterprise AI pilots fail before production. It's not because of lack of intent. People want to do this. Everybody understands the power of the technology. But why is it so hard?" The answer, according to Tallapragada, is that AI tools remain disconnected from enterprise workflows, data, and governance systems. "You're writing prompts, prompts, you're getting frustrated because the context is not there," he said, describing what he called a "prompt doom loop." Salesforce's solution is a deeply integrated platform connecting what it calls four ingredients: the Agentforce 360 agent platform, Data 360 for unified data access, Customer 360 apps containing business logic, and Slack as the "conversational interface" where humans and agents collaborate. Slack becomes the front door to Salesforce Perhaps the most significant strategic shift is the elevation of Slack -- acquired by Salesforce in 2019 for $27.7 billion -- as the primary interface for Salesforce itself. The company is effectively reimagining its traditional Lightning interface around Slack channels, where sales deals, service cases, and data insights will surface conversationally rather than through forms and dashboards. "Imagine that you maybe don't log into Salesforce, you don't see Salesforce, but it's there. It's coming to you in Slack, because that's where you're getting your work done," Harris explained. The strategy includes embedding Salesforce's Agentforce agents for sales, IT service, HR service, and analytics directly into Slack, alongside a completely rebuilt Slackbot that acts as a personal AI companion. The company is also launching "Channel Expert," an always-on agent that provides instant answers from channel conversations. To enable third-party AI tools to access Slack's conversational data, Salesforce is releasing a Real-Time Search API and Model Context Protocol server. Partners including OpenAI, Anthropic, Google, Perplexity, Writer, Dropbox, Notion, and Cursor are building agents that will live natively in Slack. "The best way to see the power of the platform is through the AI apps and agents already being built," Rob Seaman, a Salesforce executive, said during a technical briefing, citing examples of startups "achieving tens of thousands of customers that have it installed in 120 days or less." Voice and IT service take aim at new markets Beyond Slack integration, Salesforce announced major expansions into voice-based interactions and employee service. Agentforce Voice, now generally available, transforms traditional IVR systems into natural conversations that can update CRM records, trigger workflows, and seamlessly hand off to human agents. The IT Service offering represents Salesforce's most direct challenge to ServiceNow, the market leader. Mudhu Sudhakar, who joined Salesforce two months ago as senior vice president for IT and HR Service, positioned the product as a fundamental reimagining of employee support. "Legacy IT service management is very portals, forms, tickets focused, manual process," Sudhakar said. "What we had a few key tenets: conversation first and agent first, really focused on having a conversational experience for the people requesting the support and for the people providing the support." The IT Service platform includes what Salesforce describes as 25+ specialized agents and 100+ pre-built workflows and connectors that can handle everything from password resets to complex incident management. Early customers report dramatic efficiency gains Customer results suggest the approach is gaining traction. Reddit reduced average support resolution time from 8.9 minutes to 1.4 minutes -- an 84% improvement -- while deflecting 46% of cases entirely to AI agents. "This efficiency has allowed us to provide on-demand help for complex tasks and boost advertiser satisfaction scores by 20%," said John Thompson, Reddit's VP of sales strategy and operations, in a statement. Engine, a travel management company, reduced average handle time by 15%, saving over $2 million annually. OpenTable resolved 70% of restaurant and diner inquiries autonomously. And 1-800Accountant achieved a 90% case deflection rate during the critical tax week period. Salesforce's own internal deployments may be most telling. Tallapragada's customer success organization now handles 1.8 million AI-powered conversations weekly, with metrics published at help.salesforce.com showing how many agents answer versus escalating to humans. Even more significantly, Salesforce has deployed AI-powered sales development representatives to follow up on leads that would previously have gone uncontacted due to cost constraints. "Now, Agentforce has an SDR which is doing thousands of leads following up," Tallapragada explained. The company also increased proactive customer outreach by 40% by shifting staff from reactive support. The trust layer problem enterprises can't ignore Given enterprise concerns about AI reliability, Salesforce has invested heavily in what it calls the "trust layer" -- audit trails, compliance checks, and observability tools that let organizations monitor agent behavior at scale. "You should think of an agent as a human. Digital labor. You need to manage performance just like a human. And you need these audit trails," Tallapragada explained. The company encountered this challenge firsthand when its own agent deployment scaled. "When we started at Agentforce at Salesforce, we would track every message, which is great until 1,000, 3,000," Tallapragada said. "Once you have a million chats, there's no human, we cannot do it." The platform now includes "Agentforce Grid" for searching across millions of conversations to identify and fix problematic patterns. The company also introduced Agent Script, a new scripting language that allows developers to define precise guardrails and deterministic controls for agent behavior. Data infrastructure gets a major upgrade Underlying the agent capabilities is significant infrastructure investment. Salesforce's Data 360 includes "Intelligent Context," which automatically extracts structured information from unstructured content like PDFs, diagrams, and flowcharts using what the company describes as "AI-powered unstructured data pipelines." The company is also collaborating with Databricks, dbt Labs, and Snowflake on the "Universal Semantic Interchange," an attempt to standardize how different platforms define business metrics. The pending $8 billion acquisition of Informatica, expected to close soon, will expand metadata management capabilities across the enterprise. The competitive landscape keeps intensifying Salesforce's aggressive AI agent push comes as virtually every major enterprise software vendor pursues similar strategies. Microsoft has embedded Copilot across its product line, Google offers agent capabilities through Vertex AI and Gemini, and ServiceNow has launched its own agentic offerings. When asked how Salesforce's announcement compared to OpenAI's recent releases, Tallapragada emphasized that customers will use multiple AI tools simultaneously. "Most of the time I'm seeing they're using OpenAI, they're using Gemini, they're using Anthropic, just like Salesforce, we use all three," he said. The real differentiation, executives argued, lies not in the AI models but in the integration with business processes and data. Harris framed the competition in terms familiar from Salesforce's founding: "26 years ago, we just said, let's make Salesforce automation as easy as buying a book on Amazon.com. We're doing that same thing. We want to make agentic AI as easy as buying a book on Amazon." The company's customer success stories are impressive but remain a small fraction of its customer base. With 150,000 Salesforce customers and one million Slack customers, the 12,000 Agentforce deployments represent roughly 8% penetration -- strong for a one-year-old product line, but hardly ubiquitous. The company's stock, down roughly 28% year to date with a Relative Strength rating of just 15, suggests investors remain skeptical. This week's Dreamforce demonstrations -- and the months of customer deployments that follow -- will begin to provide answers to whether Salesforce can finally move enterprise AI from pilots to production at scale, or whether the "$7 billion business" remains more aspiration than reality.

[7]

Axios

Exclusive: Salesforce announces agents with voice and hybrid reasoning

Why it matters: Salesforce faces growing competition from tech giants with their own large language models and startups focused on AI customer service. Zoom in: Salesforce agents will now have the power of speech and new hybrid reasoning skills. * The upgrades aim to make them sound more human while following clear rules for key decisions, like giving refunds. Driving the news: The features will be part of a flurry of announcements that Salesforce will make at its annual Dreamforce event next week. * Agentforce Voice works across phone systems, websites and apps, and will be widely available on Oct. 21. * Hybrid reasoning will debut as part of a feature called Agent Script, now in pilot, with a customer beta coming in November. What they're saying: Both features reflect that there's no one-size-fits-all approach to AI agents, Salesforce Executive VP Adam Evans told Axios Wednesday. * "You can have the creativity and fluidity when you want it, or you can have the rigidity, consistency and scale when you don't," he said. "It's your choice on a job-by-job basis." * Companies appreciate that bots don't sound like bots, but they also need them to stick to a consistent approach, Evans said. * "If you're doing 100,000 transactions with the customer and some of them are different than the others, that's not a feature," Evans said. "That's a bug, right?" Salesforce said it's spent about a year developing the new voice features, focusing on understanding speech and detecting nuance and emotion to respond appropriately. * The voice system can also pull data from various systems and seamlessly hand off to a human worker when needed. * "There's not very many companies that can do that," Evans said. Enterprise agents for customer service is a saturated space with fierce competition from tech companies like ServiceNow and startups like Sierra, which was co-founded by Bret Taylor, a former top Salesforce executive.

[8]

Salesforce's Agentforce software is coming to OpenAI's ChatGPT later this year

Salesforce CEO Marc Benioff participates in an interview during the World Economic Forum in Davos, Switzerland, on Jan. 22, 2025. Salesforce is ramping up partnerships with leaders in generative artificial intelligence as investors continue to fear that the software company faces business risks due to the rapid growth of AI. Just ahead of its annual Dreamforce conference in San Francisco, Salesforce said Tuesday it will enable the use of AI models from OpenAI and Anthropic inside its Agentforce 360 software. A day earlier, Salesforce expanded Agentforce beyond text chats to also handle voice calls. "The way people are going to interact with software is going to fundamentally shift," said Brian Landsman, CEO of Salesforce's AppExchange business and executive vice president of partnerships, in an interview. The interaction could be in ChatGPT or in Slack, he said. Salesforce will collaborate with Anthropic to bring Agentforce 360 into Claude, Landsman added. Shares of Salesforce are down about 26% this year, while the S&P 500 index has gained 13%, as Wall Street seeks faster revenue growth from the cloud software company. So far, Agentforce revenue has been "modest," Morgan Stanley analysts, who have the equivalent of a buy rating on Salesforce, wrote in a Monday note. Large software companies are increasingly turning to popular AI model developers for new capabilities. Atlassian, Datadog and Intuit have previously signed deals with OpenAI, and Microsoft has invested almost $14 billion in the company. In September, Databricks committed to spending $100 million on OpenAI models. As part of Salesforce's announcement, customers will be able to access corporate information in Agentforce 360 and create charts in Tableau through the ChatGPT assistant, which has more than 800 million weekly users. Last week OpenAI announced a software development kit for integrating third-party applications into ChatGPT. Companies working with both OpenAI and Salesforce will be able to sell products through ChatGPT's instant checkout feature later in 2025. Salesforce plans to work with Anthropic on selling products for regulated industries, starting with financial services. OpenAI said last month that ChatGPT users would be able to purchase products from U.S. Etsy sellers and Shopify merchants. Meanwhile, Salesforce said its engineering organization is adopting Anthropic's Claude Code programming product. "We plan to continue to go much deeper with these partners over time," Landsman said. Salesforce CEO Marc Benioff has been defending his company's position in the AI boom. And on last month's earnings call, he said Anthropic and OpenAI both use Salesforce tools. "All these next-generation AI companies ranging from OpenAI to Anthropic to everyone are on Slack," Benioff, who is also Salesforce's co-founder, told analysts. "And it is incredible how they've used that as their operating system and as their platform to run their companies."

[9]

TechRadar

Salesforce Agentforce 360 looks to take your company data to the next level

New addition will help customers adopt the "Agentic Enterprise" Salesforce has announced Agentforce 360, a new platform designed to help businesses adopt the "Agentic Enterprise". Covering everything from sales leads to IT tickets, Agentforce 360 uses new levels of AI to improve efficiency and the overall customer experience using the latest agentic technology. "We've gone faster with Agentforce than we've gone with any of our other technology," said Marc Benioff, Salesforce CEO and founder, announcing the launch at the company's Dreamforce 2025 event in San Francisco - a year since the initial release of Agentforce as a product. The Agentforce 360 Platform is the latest release of Agentforce, designed for building, controlling, and deploying enterprise AI agents at scale. It aims to provide controlled, contextual, and consistent experiences for customer and employee interactions, accelerating the path from AI prototypes to production-scale agents. It will provide an easier way for companies to build agents and rapidly deploy them across their workforce using a conversational interface, hopefully allowing users to build exactly the services they require. Working alongside Salesforce's other platforms - Data 360, Customers 360 and Slack, Salesforce hopes the launch will help customers lower costs and become more productive, as it also integrates with a host of third-party tools. "This is the next revolution...we've gone through predictive AI, and we're now entering this new agentic AI revolution," Benioff added.

[10]

Dreamforce 25 - AI, meet HR! Salesforce Chief People Officer Nathalie Scardino explains how agents change everything

Be careful what you wish for! For as long as I've been in and around the enterprise tech game, I've heard a common lament from Chief HR Officers (CHROs) across industries, namely that if only they didn't have to focus on routine process and paperwork and admin minutiae as a core part of the day job, then they could be sitting at the boardroom table making a more strategic contribution to the overall growth of the business. Well, the rise of agentic tech might well be calling HR's bluff in this respect. If an agent can take all those mundane routine tasks off your hands, then that does indeed finally set you free to shine, doesn't it? At which point the question becomes, are you and your peers actually ready in practice to step up to the mark to deliver on what you've been promising impotently for so long? If you check out diginomica's latest Executive Intelligence podcast, with Greg Shewmaker, CEO of advisory firm r.Potential, his blunt assessment is that HR isn't ready for what the agentic revolution has to offer. On the other hand, Salesforce's Chief People Officer Nathalie Scardino appears more enthused when I put the question to her: Yesterday, for the first time at Dreamforce, we had a CHRO summit, because the role of HR now is to remove that bureaucracy, and it starts in your own organization. For us, we moved more to kind of this product mindset, where everything that we produce, whether it's Employee 360, Career Connect, is about removing that friction. I think HR has this massive opportunity to do that in a way that maybe it hasn't before because of the emergence of this new workforce transformation. It's why I feel so deeply about the role of the HR executive in in the whole agent/human collaborative workforce. It is a duty, I think, of HR, to make sure that they're leading from the front. And I do think, from the summit that we had yesterday, everybody is thinking this. Every role has a level of bureaucracy to it, of paper-pushing, that should be eradicated immediately, because who has time for that? We have to focus on what's the most important, and that's driving growth and innovation and for us being the best place to work, so retaining the best and brightest. Of course this does then beg the question of how enterprises should introduce agentic tech into the HR process and the whole issue of adoption by or resistance from users? There are precedents here involving other new tech, suggests Scardino: It wasn't like we came into work and said, 'Hey, who wants to use email?'. It's not a question, right? It's a transformation that you have to do and it's the same on the AI front. Every single job, all of us, every job is being re-designed. We have a workforce innovation team, and we did a lot of research with the World Economic Forum, with our partners at LinkedIn, and really thought about a framework that we call the four R's. The first R is re-design, she explains: Every job, fundamentally at the task level, is re-designed with AI. And then you have to consider, once you understand that baseline, how do you re-skill? That has been our huge push internally. Publicly, we've talked about re-skilling a million Agentblazers. Everybody in our company is an Agentblazer champion, but you couldn't get an invite to our last executive management meeting without being...hands on, keyboard building agents. So re-skilling is this constant, learning being the meta skill for everybody. Next up is re-deployment: A lot of work roles are starting to merge. There are synergies between skill sets because of the access to technology, because the agent is taking an action. Finally there's re-deploying your talent and re-balancing within a digital workforce as the final elements of the framework. There are some key learnings that have emerged from all this, says Scardino: We know that the more high-performing teams, the most innovative teams also, are the most engaged. What drives that engagement is that they have access to tools that they test, that they iterate. So there's a direct correlation between whether your manager is leveraging AI and your engagement and your trust, because it's the ambiguity where the disconnect comes in. The implications for introducing AI adoption into HR are clear: It's not like we're instantly going live with everything. We work really closely with our Chief Digital Officer and the Office of Ethical and Humane Use of Technology, and we figure out, where do we want to test, where do we want to pilot, and what is the feedback based on on the experience? In fact, Scardino argus that HR has fundamentally changed already: I now have a Workforce Innovation team that I didn't have 12 months ago on the consciousness around ethics. I actually see that as [being] an 'everybody responsibility'. We have this Governance Council around, just because you can, should you? She cites the example of recruitment as a case in point: We have an agent in the recruiting function, but it is looking at all of our job applications. It is not making the hiring decision. We had this core principle three years ago that we would never leverage AI to make a hiring decision. So I think having these principles up front [is important]. What are the non-negotiables? Non-negotiables are the ethical use and deployment of AI for good. As to fears about the potential negative impact on jobs of AI, Scardino argues that with any technological advancement, there is a period of uncertaintyL There was a time when spreadsheets came out, we thought there'd be an eradication of accountants. It's the opposite. We have more of them, but they're doing higher value work. So for us, the first thing began around everybody being an Agentblazer, top down, bottoms up. It doesn't matter your level or function, everybody is required to gain these AI fluency skills from a accreditation standpoint. Irrespective of your role, understanding the data architecture is a core skill, no matter what function you're in. On skills, what we determined about eight months ago is we align around our top 10 enterprise skills, where every single employee would have access to tools like Trailhead through to hands on keyboard learning. And AI will also bring about new roles to manage, she argues: There are net new jobs that we have today that we didn't have even six months ago. You know, Ethical AI Architects, Prompt Engineers, Agent Manager, Supervisors in Conversational UI design, Agent Designers. Talent Management in my own org is now somebody that's supporting agent and human workforce collaboration. We have business skills. You have the very human skills - self awareness, emotional intelligence, creativity - and then agent skills, which is your AI literacy and your data literacy. So with Career Connect, which is our internal talent marketplace, we have focused aggressively on ensuring that every single employee has access to these Top 10 skills. My take Change is the only constant and what seems like sound thinking at one point can rapidly come to seem out of date. Scardino jokes: My business dissertation in 1999 was on whether the Body Shop PLC should go online...and I thought I was cutting edge then. The role of HR in the agentic age is an evolving debate. By the time Dreamforce 2026 comes around, everything may have moved on again as adoption levels increase and the realities of managing a Digital Labor workforce kick in.

[11]

Agentic orchestration: Salesforce's new company roadmap - SiliconANGLE

Agentic orchestration takes center stage as Salesforce evolves beyond CRM Agentic orchestration is emerging as the connective tissue of the modern enterprise -- where intelligent agents, data and workflows converge to power adaptive, self-optimizing systems. As organizations shift from static automation to dynamic collaboration between artificial intelligence-driven agents, the foundation of how businesses operate and innovate is being fundamentally rewritten. In this new era of distributed intelligence, Salesforce Inc. is positioning itself at the center of that transformation. Its latest vision goes beyond customer relationship management to create a platform capable of orchestrating agents, data and workflows in concert. The goal is not just to manage customer experiences, but to intelligently shape them -- anticipating needs, coordinating outcomes and driving value at unprecedented scale, according to Holger Mueller (pictured, left), vice president and principal analyst at Constellation Research Inc. "We all know agentic AI is only as good as the APIs in the background," Mueller said. "Pretty much all the vendors are acting like it's all about AI and we don't have to upgrade the background business processes ... that was a big miss from my perspective." Mueller and Raphaelle d'Ornano (right), founder and chief executive officer of Decoding Discontinuity, spoke with theCUBE's John Furrier and Gemma Allen at Dreamforce, during an exclusive broadcast on theCUBE, SiliconANGLE Media's livestreaming studio. They discussed Salesforce's new focus on AI and agentic orchestration and what it means for the enterprise. (* Disclosure below.) Salesforce's decision to expand its product purview only underscores long-underway internal changes, according to d'Ornano. Having overgrown its restrictive identity in CRM ages ago, such a directional shift cements the company's maturity and ability to pivot toward expansion. "My first take was that they should change their ticker," she said. "It makes no sense that Salesforce is CRM anymore because that's just not what they are at all. I think, when you come out of this keynote, you say to yourself, 'Salesforce has perfectly understood that agentic AI is a new paradigm and that's been clear from day one.' Today, Salesforce is, for me, an agentic AI orchestration platform, able to go not just in the CRM space, but to the broader enterprise orchestration landscape and the names that we have all over here of the agentic enterprise, that's a very good qualification term for Salesforce." The broader signal is Salesforce positioning itself to redefine software categories altogether, d'Ornano noted. Traditional boundaries between infrastructure, horizontal applications and business platforms no longer apply. The company's strength lies in orchestrating data, workflows and context -- positioning itself as the central nervous system of the agentic enterprise. "Enterprise AI adoption and agentic AI adoption is something that is an architectural redefinition of a business," d'Ornano said. "You're changing the architecture, your tech architecture, you're changing your business model architecture. By doing that, you're changing the business model of your clients' architecture. This is not as quick as the adoption of ChatGPT. I would also add that they're actually in a very good position in the sense that they've mastered customer data ... they're actually super well-positioned to build that orchestration platform." Execution will be crucial at such a critical juncture for Salesforce, both panelists agreed. The company's control of customer data -- a rich and underexploited asset -- gives it an advantage in building agentic orchestration platforms. The question now is whether it can turn that data mastery into differentiated agentic outcomes at scale. Here's the complete video interview, part of SiliconANGLE's and theCUBE's coverage of Dreamforce:

[12]

Analytics India Magazine

Salesforce, OpenAI Partner to Integrate Agentforce 360 and ChatGPT | AIM

Businesses will also be able to use OpenAI's latest frontier models, including GPT-5, to create AI agents and prompts within the Salesforce Platform. Salesforce and OpenAI have expanded their strategic partnership to integrate Salesforce's Agentforce 360 platform and OpenAI's frontier models across ChatGPT, Slack, and enterprise workflows, the companies announced on Tuesday. The collaboration will allow users to access Salesforce's Agentforce 360 directly in ChatGPT, where they can query sales records, review customer conversations, and build Tableau visualisations using natural language prompts. Salesforce said more details on the availability of Agentforce 360 apps and Agentforce Commerce in ChatGPT will be announced in the coming months. ChatGPT and Codex integrations in Slack are available now. Businesses will also be able to use OpenAI's latest frontier models, including GPT-5, to create AI agents and prompts within the Salesforce Platform. A new commerce experience will extend to ChatGPT, allowing merchants using Agentforce Commerce to reach millions of potential US customers through Instant Checkout while maintaining control over their data and relationships. "Our partnership with Salesforce is about making the tools people use every day work better together, so work feels more natural and connected," said Sam Altman, co-founder and CEO of OpenAI. "It's an important step in how AI can improve daily workflows under our efforts together." Salesforce chair and CEO Marc Benioff said the partnership brings enterprise intelligence closer to real-time decision-making. "By uniting the world's leading frontier AI with the world's #1 AI CRM, we're creating the trusted foundation for companies to become Agentic Enterprises," Benioff said. The integrations announced as part of the partnership include Agentforce 360 apps in ChatGPT, which bring CRM data and Tableau visualisations directly into ChatGPT conversations. Agentforce Commerce will connect with Instant Checkout, powered by the Agentic Commerce Protocol, to support transactions and order management directly within ChatGPT. The partnership also extends to Slack, where ChatGPT and Codex integrations will enable teams to summarise conversations, draft content, and automate coding tasks through natural language commands. Additionally, OpenAI's models will be available within the Agentforce 360 platform, offering reasoning, voice, and multimodal capabilities as preferred options for the Atlas Reasoning Engine and Prompt Builder. The integration also extends to commerce. Businesses using Agentforce Commerce can list products in ChatGPT, where consumers can browse and complete purchases using Instant Checkout with Stripe. The process maintains privacy compliance and keeps merchants in control of customer relationships.

[13]

Salesforce adds voice calling to Agentforce AI customer service software

Salesforce is adding voice to its Agentforce software, letting clients go beyond text when using artificial intelligence agents to respond to customer questions. With Agentforce Voice, companies can customize the tone and speed of voices and adjust the pronunciation of specific terms, Salesforce said Monday, ahead of its Dreamforce conference in San Francisco this week. The feature also allows people to interrupt the AI agent during phone calls. Voice is becoming a bigger part of the generative AI boom, which started with text-based prompts in late 2022, when OpenAI launched ChatGPT. In the past year, OpenAI and Anthropic have enabled their chatbots to conduct spoken conversations without sounding overly robotic. Now that capability is taking hold inside business software. Agentforce Voice will integrate with corporate phone systems from Amazon, Five9, Genesys, Nice and Ericsson's Vonage. Former Salesforce co-CEO Bret Taylor is also trying his hand in the market. Taylor helped start Sierra in 2023, and last year the startup announced that its AI agents "can now pick up the phone." Sierra has been valued at $10 billion, and has a client list that includes ADT, SiriusXM and SoFi. Salesforce has been under pressure this year in part due to investor concern that software companies could lose business as AI moves deeper into coding. The stock is down about 28% so far in 2025, while the Nasdaq has gained around 15% over that stretch. Anthropic told reporters in September that its Claude Sonnet 4.5 model built a chat app similar to Salesforce's Slack in 30 hours. In Salesforce's latest earnings report, the company warned that new AI products "may disrupt workforce needs and negatively impact demand for our offerings." Salesforce CEO Marc Benioff has downplayed the risk to this company. "When we get into this kind of zero-sum game, well, all this is going to get wiped out, or all this is going to change, then, you know, you're not dealing with somebody who actually runs a company, because that's not the way business works," Benioff told CNBC's Morgan Brennan last month. "Business is incremental, it's evolutionary, it's growing, it's evolving, and we don't see that kind of change." Salesforce launched Agentforce last year as a service that could respond to customer requests over text chats with help from generative AI models. Agentforce now has more than 12,000 implementations, according to a statement. But there's some skepticism about its popularity. "Investor enthusiasm around Agentforce has moderated as adoption has lagged expectations," RBC Capital Markets analysts, who recommend holding the stock, wrote in a note to clients last week. In November, Salesforce will provide early access to Agent Script software, which organizations can use to customize what agents say and do.

[14]

Dataconomy

Salesforce launches Agentforce 360 to supercharge enterprise AI agents

The new platform integrates with Slack, allowing Sales, IT, and HR AI agents to appear directly in chats. Slack will soon feature a reworked Slackbot that acts as a personalized, learning AI assistant. Salesforce has announced Agentforce 360, an updated AI agent platform, ahead of its October 14 Dreamforce conference. The new version introduces enhanced tools for building and instructing agents, aimed at attracting enterprise clients in a competitive market. A central feature of Agentforce 360 is Agent Script, a new prompting tool entering a beta phase in November. It allows users to program agents with "if/then" logic to produce more predictable and flexible responses in scenarios like customer questions. To support this, users can access "reasoning" models designed to process information before responding, as opposed to relying on pattern recognition. These models are powered by technology from Anthropic, OpenAI, and Google Gemini. Salesforce is also releasing Agentforce Builder, a new tool for constructing, testing, and deploying AI agents from a single platform. Scheduled for a November beta release, this tool incorporates Agentforce Vibes, which the company previously announced as an enterprise-grade app-vibe coding tool. A broader integration between Agentforce and the Slack messaging application was also announced. Core Agentforce applications for Sales, IT, and HR will begin to surface directly within Slack starting this month, with the integration expanding through the beginning of 2026. Concurrently, Slack is piloting a new version of its Slackbot. This updated bot is designed as a personalized AI agent that learns from user activity to offer insights. Salesforce also intends for Slack to serve as an enterprise search tool, with connectors for Gmail, Outlook, and Dropbox planned for release in early 2026. Salesforce reports that its Agentforce platform has 12,000 customers. In its press release, the company stated this figure is "significantly higher than any of its competitors." Early pilot customers for the Agentforce 360 upgrades include Lennar, Adecco, and Pearson. The update arrives as numerous companies introduce AI features for enterprise customers, while these enterprises report difficulties achieving a return on investment. This announcement follows recent activity from competitors. Last week, Google announced Gemini Enterprise, a suite of tools for building enterprise-grade AI agents, naming Figma, Klarna, and Virgin Voyages as early customers. Anthropic also reported progress with its Claude Enterprise product, securing a contract with Deloitte to provide its Claude chatbot to 500,000 global employees, its largest enterprise deal yet. The day after this deal was publicized, Anthropic announced a strategic partnership with IBM.

[15]

Inc. Magazine

This New Salesforce AI Service Could Cut IT Helpdesk Calls at Your Company

Workplace software services giant Salesforce just revealed its Agentforce IT Service product -- an AI agents-based system that offers always on, always available IT support and helplines for its client companies. In a Thursday press release, Salesforce reiterated some of the same arguments made by AI evangelists, promising its use could allow workers and IT staff alike to "spend less time on manual tickets, forms, portals, and searching through knowledge articles, and more time on high-value, strategic work." Even if your company isn't one of Salesforce's myriad clients, this system is a little sample of the future, and yet another example of how AI is encroaching on diverse sectors of everyday work life. There's one caveat to the new tool, though. It still relies on people. We've all been there: sending a "help!" email to the IT team because some important piece of software or hardware has gone kaput, only to receive a ticket number or case number from an automated system, usually accompanied by a note saying the team will respond "soon," or, worse, describing a longish wait window like "by tomorrow." Depending on the setup, and the nature of the problem someone will then show up in your office, or send suggested fixes (which may read like a foreign language to a non-technical worker) by email or a messaging app, or take over someone's computer via remote access to fix the problem. The Salesforce tool, the company insists, is unique because it's "conversation-first," and "agent-first." Essentially the idea is to dump the "ticket" system, and allow someone to make an IT help request via pretty much any platform they're using, from chat systems like Slack or Teams to email. In a demonstration press event, one example featured a new employee who needed to go through their IT onboarding. They began a chat in Slack with an AI "conversation agent," which verified a few details with the worker, then set to work sending them the relevant documents and guidelines, as well as actually working behind the scenes to, for example, give the worker access to file systems, GitHub code repositories and other information for their onboarding. It's able to do this because unlike a query-then-response AI chatbot, agent AIs have a degree of autonomy and can perform some digital tasks automatically. It sounds like magic, and unlike, say, having a long wait for an IT operative in a remote call center, the system is effectively on 24/7/365. Salesforce also demonstrated that the AI agent system also works in a similar way for IT support people as well, offering answers to technical problems via a chat interface. IT-savvy readers, or perhaps IT-wary ones, may have some worries at this point. It's one thing to trust a human expert with your computer when, say, an important Excel file you've worked on for hours gets corrupted. But AI systems aren't human, and we know that they can hallucinate fake or incorrect outputs and pass them off as true or reliable. When Inc. asked Salesforce about this during a press conference about the new agentic AI IT service, Muddu Sudhakar, Salesforce's senior vice president and general manager of IT Service and HR Service agreed this was the "most important" question concerning AI deployments today. Then he said the company's multiple AI agents were trained carefully, and operated within "guardrails" that should keep them in line, and prevent serious errors occurring if, for example, the AI suggested a fix for a user's computer that actually makes things worse. Salesforce also noted that there's always a "human in the loop" as part of this AI-centric system. Someone who should be able to spot if an AI has made an error, or to whom you can "escalate" the issue if you're not confident the AI can fix your problem. What can you take away from this for your company? First, this is a hint of the future. Not just for IT services, but for many business support systems that are likely to operate like this as AI chatbots and AI agents get more powerful. If you contract out to third-party companies for, say, IT or financial service support, then it's likely that you'll be interacting with AI agent-based systems soon, instead of humans first. Salesforce has previously released an AI agent product that can work like a sales rep -- so you see which way the wind is blowing. Second, Salesforce's human-in-the-loop model is a reminder that while AI tools can boost worker efficiency, they are not perfect and they can make mistakes. (The Salesforce IT model speeds the process, so, the IT team may be able to deal with incoming user queries faster after AI tools do a big share of the initial work.) If you've rolled out AI at your business, you should remind your workers that every AI output needs to be checked for veracity and relevance before it's built into any product -- that way you can avoid problematic, or even legal, expensive, AI-induced mistakes.

[16]

Dreamforce 25 - Certinia's COO on transparency, strategic discipline, and building trust

Among the packed schedule of sessions and announcements at Dreamforce 2025, Certinia's Chief Operating Officer Robert Cesafsky offered something particularly valuable: operational honesty. Over the course of a conversation during the event, Cesafsky revealed an organization building its AI strategy on a foundation of transparency, measured investment, and strategic clarity about keeping a complex product portfolio aligned. When a new leadership team joined in late 2023, the company defined its position around managing "customers, projects, people and financials across the entire customer lifecycle journey" - a statement that now anchors every strategic decision. Those three applications - PS Cloud (Professional Services Automation), CS Cloud (Customer Success Management), and FM Cloud (Financial Management) - can be deployed independently or together, giving customers flexibility while maintaining platform coherence. According to Cesafsky, this clarity has fundamentally reshaped how the company operates, from individual product roadmaps to customer support and partner ecosystem strategies. For Cesafsky, the definitional statement is more than marketing language - it has real operational implications. As he said to his team during the strategic planning process: I promise you it's not a marketing exercise. There's actually real world implications around this. The challenge is ensuring that three distinct product lines evolve coherently while each serves its own market. Cesafsky describes a disciplined approach of reviewing overall strategy and where growth comes from, conducting risk assessments within the existing customer base, and rebalancing R&D investments annually based on strategic imperatives. This isn't always straightforward. During the first year of repositioning, the leadership team recognized that Financial Management Cloud customers needed clearer signals about the solution's strategic importance. He recalls: I think we had take stock and say, 'This is actually a core, strategic part of our solution set. We need to be investing into it. We need to be communicating that to both our customers and our prospects that it is a strategic part of our product portfolio'. The company makes changes based on customer feedback and market dynamics, but always through the lens of that unifying statement - a framework that forces alignment across product roadmaps, go-to-market strategies, and support operations. When it comes to AI specifically, Certinia has taken a systematic approach. The company currently has two agents on Salesforce's Agent Exchange platform, complementing generative AI capabilities already embedded in its core solutions. But the most intriguing aspect is how concrete the roadmap is. The company has identified 164 unique tasks and actions performed by key personas: resource managers, service delivery managers, project managers, and consultants. Cesafsky explains: Our AI and agentic roadmap is actually quite easy. Think about what your users are doing today, and we want to more fully automate those. Each of these 164 tasks ties back to specific ROI drivers - improvements in services win rates, more efficient resource management, higher billable utilization, and net reduction in systems and personnel used for resource and project management. It's a framework that makes the roadmap measurable rather than aspirational. Importantly, Cesafsky is explicit about what Certinia is not doing with AI: We're not going to be the company that stands up and proclaims that we were able to reduce headcount by 20% because we're deploying agents right now. The focus remains on making professional services people more effective, not eliminating them. This also reflects a broader commitment to accountability. The company has developed a "Customer Success Scorecard" tracking ROI drivers across its 1,400+ customer base. But Cesafsky emphasizes willingness to acknowledge when promised outcomes aren't materializing. He notes that on the occasion when real-time data has revealed gaps between some longstanding ROI claims and actual customer results, the company either fixes the capability or stops claiming it - a level of honesty that connects to Cesafsky's principles about building customer trust. For a 17-year-old company, managing stakeholder expectations while evolving presents unique challenges - and this is where Cesafsky's perspective becomes most revealing. Certinia began as FinancialForce, developing a General Ledger on Salesforce's platform. Over nearly two decades it has pivoted multiple times, and when stakeholders joined the ecosystem significantly influences their perception of what Certinia is. Cesafsky explains: If you joined as a customer in 2010 you very much think about us as the finance system built on Salesforce. If you joined as a partner in 2017 your perspective is something around they want to be the business services ERP vendor. Each era created a different mental model, and stakeholders naturally anchor to the version of the company they first encountered. Because each phase leaves its mark, he adds, that puts a responsibility on leadership: Our job as leaders, our job as a vendor, is just to continually communicate as effectively, as clearly, as frequently as we can to all of those stakeholders within Certinia about where we are and where we're going, When asked about maintaining trust, Cesafsky's response was unequivocal: It's imperative. There has to be trust. There has to be transparency with all of our people, both internally and with our customers as well. What emerges from this conversation is leadership that understands trust must be earned through transparency, not announcements. While plenty of vendors rush to claim AI transformation while struggling to articulate coherent platform strategies, Certinia's leadership is building something more durable - a culture of operational honesty paired with strategic clarity. The details matter. The definitional statement that forces alignment across three products. The 164-task framework that makes automation concrete. The explicit rejection of headcount reduction as the primary AI goal. The willingness to course-correct as necessary with both investment and communication. Cesafsky's approach reveals something crucial - customers don't need more AI announcements, they need vendors who understand how to keep complex product portfolios aligned while being honest about what works, what doesn't, and what they're still figuring out.

[17]

Inside Agentforce's new era for the AI development lifecycle - SiliconANGLE

Agent-driven software steps out of the lab and into production Code is entering a new creative era -- one where the development lifecycle itself is becoming a living, adaptive system. Engineers are no longer just writing code; they're orchestrating an evolving interplay between data, logic and automation that redefines how software grows and improves over time. As agentic systems mature from prototypes to production-ready tools, development is shifting from speed to sophistication. The focus now lies in crafting workflows that mirror human reasoning, ensuring systems can learn, adapt and act with purpose. It's a reinvention of engineering culture -- merging experimentation with precision and transforming the once-linear process of software delivery into an ongoing dialogue between human intent and computational design, according to Jayesh Govindarajan (pictured, right), executive vice president of software engineering at Salesforce Inc. "We are seeing the birth of a new development life cycle, an agent development life cycle ... and pretty much all the tools that we are building around it -- a studio for creating agents, a testing center for testing them, an observability command center for seeing fleets of agency in action and then being able to run that loop to make them better over time; that's the iterative cycle," Govindarajan told theCUBE. "Developers are operating at a much higher level than before. Don't make the mistake of thinking that this is the same work that used to happen before." Govindarajan and Christophe Coenraets (left), senior vice president of technical audience relations at Salesforce, spoke with theCUBE's John Furrier and George Gilbert at Dreamforce, during an exclusive broadcast on theCUBE, SiliconANGLE Media's livestreaming studio. They discussed how Agentforce is redefining AI-driven software creation, enterprise vibe coding and the craftsmanship behind modern development. (* Disclosure below.) Agentforce's growing adoption highlights the industry's hunger for tools that streamline production without losing rigor. The platform now serves about 12,000 customers, half of whom are paid users, according to Govindarajan. Each deployment has taught new lessons in reliability, determinism and creative flexibility -- the qualities that distinguish demos from production-ready agents. "If you go to Williams-Sonoma today, you can go onto their website and there's a sous-chef agent, which is basically built on Agentforce," he said. "You're going from a creative process to a more lockdown process, which is now [that] you're exchanging monies, you're going from insight to commerce." The creative-to-deterministic balance is at the core of enterprise adoption. Developers must manage when an agent can improvise and when it must follow strict rules. That tension -- between openness and structure -- is what Salesforce calls "vibe coding," according to Coenraets. "Vibe coding is going to give you the quick prototype, but in the enterprise, developers want security, they want trust, they want governance, and they also want the code to be grounded in the metadata that they have and all the security features. Enterprise vibe coding is really filling that gap, and what we are showing here with Agentforce Vibes is exactly that: Vibe coding for the enterprise." AI's influence on the development lifecycle extends beyond the code itself. Developers are discovering new ways to design user experiences around unpredictable agent behavior. This innovation means preparing for tasks that may not have existed at design time -- a catalyst for conversational user experience to become the new default in enterprise apps, Coenraets explained. As enterprises continue refining their AI strategies, the development lifecycle is becoming a blueprint for how creativity and control can coexist at scale. "Agentic AI [is] completely changing how you build software," he said. "IT shops have incredible backlogs. I think what you're seeing is that because the productivity is much higher, it means more software is being built, and I think developers are feeling that." Here's the complete video interview, part of theCUBE's coverage of Dreamforce:

[18]

Analytics India Magazine

Salesforce Pledges $15 Billion Investment in San Francisco, Launches Agentforce 360 | AIM

According to Salesforce, more than 12,000 customers are already using Agentforce 360. Salesforce on Monday announced plans to invest $15 billion in San Francisco over the next five years to support AI innovation, workforce development, and company transformation initiatives. The investment includes a new AI Incubator Hub on Salesforce's San Francisco campus and funding for workforce training programs. Salesforce said the initiative will help companies transform into "Agentic Enterprises," a framework the company developed to integrate AI into workplace operations. Besides that, the company also announced the global launch of Agentforce 360, a new platform that connects humans and AI agents in one trusted system. The announcement, made at Dreamforce 2025, introduces what the company calls the "Agentic Enterprise", a new model where AI works alongside people rather than replacing them. "We're entering the age of the Agentic Enterprise -- where AI elevates human potential like never before," said Marc Benioff, chair and CEO of Salesforce. "Agentforce 360 connects humans, agents, and data on one trusted platform, helping every employee and every company achieve more than they ever thought possible." Salesforce said Agentforce 360 is available globally starting today, with additional features set to roll out in pilot and beta programs in the coming months. The launch follows a year of rapid development and deployment across thousands of customer environments. Over the past 12 months, Salesforce has released four major versions of Agentforce, each expanding the platform's ability to integrate AI into enterprise workflows. With this launch, Salesforce extends beyond its traditional CRM roots. Agentforce 360 is built on four layers, including the Agentforce 360 AI foundation, the Data 360 unified data layer, the Customer 360 suite of business applications, and Slack as the collaborative interface. According to Salesforce, more than 12,000 customers are already using Agentforce 360. Early adopters have reported measurable gains in efficiency, customer satisfaction, and cost savings. The Agentforce Builder allows teams to design and deploy agents using natural language without manual configuration. A new Agentforce Voice layer enables real-time speech recognition and synthesis for customer interactions. The system also combines deterministic workflows with large language model reasoning through its Hybrid Reasoning and Agent Script tools. To ensure transparency, Salesforce added observability dashboards to monitor reasoning, accuracy, and compliance. The Data 360 layer now activates both structured and unstructured data, enabling AI agents to provide deeper business context. Through Intelligent Context and Tableau Semantics, data from across clouds and formats can be translated into consistent business language. Partnerships with Databricks, dbt Labs, and Snowflake extend this semantic framework across platforms. Slack, now central to Salesforce's AI vision, becomes what the company calls the "Agentic OS" -- the space where humans, agents, and applications work together. New Slack-first apps for sales, IT, and analytics allow employees to perform tasks and access insights directly within the chat interface. Features such as Enterprise Search and Channel Expert Agent provide real-time knowledge retrieval, while the Model Context Protocol (MCP) connects external AI providers, including Anthropic, Google, OpenAI, and Perplexity. Salesforce also launched AgentExchange, a Slack-native marketplace where companies can discover and deploy partner-built agents and integrations.

[19]

Economic Times

Salesforce deepens AI ties with OpenAI, Anthropic to power Agentforce platform

The deals, announced on Tuesday, will embed OpenAI's latest GPT-5 model and Anthropic's Claude family of models directly into Salesforce's ecosystem, enabling employees and consumers to interact with customer data and analytics in ChatGPT, Slack and Salesforce's own software. Salesforce expanded partnerships with OpenAI and Anthropic to integrate their frontier AI models into its Agentforce 360 platform, aiming to deliver enterprise-grade AI tools to a wider range of businesses and regulated industries. The deals, announced on Tuesday, will embed OpenAI's latest GPT-5 model and Anthropic's Claude family of models directly into Salesforce's ecosystem, enabling employees and consumers to interact with customer data and analytics in ChatGPT, Slack and Salesforce's own software. The twin deals underscore Salesforce's push to make its Agentforce 360 platform a central access point for major AI models, as enterprise software makers race to integrate generative AI tools into everyday business workflows. Agentforce 360, launched globally on Monday, was designed to create, deploy and manage AI agents across an entire organization. The expanded OpenAI partnership allows users to access Salesforce's Agentforce 360 directly inside ChatGPT, work with data, build Tableau visualizations and to develop AI agents within the Salesforce platform. A new "Agentforce Commerce" feature will also let merchants sell products via ChatGPT's Instant Checkout while retaining control of customer data and fulfillment. Anthropic's Claude models will be available to customers in regulated sectors such as financial services, healthcare and cybersecurity through Salesforce's secure cloud environment. The companies will also develop industry-specific AI tools and integrate Claude more deeply into Slack and the Agentforce 360 platform.

[20]

Dreamforce 2025 - the importance of human intelligence in the agentic mix, according to Aum Lifetech

Agentic AI technology can free up human beings from repetitive tasks that can be automated, leaving people free to apply their creativity to bigger challenges facing humanity. That's the view expressed by Veenu Aishwarya, CEO of Aum Lifetech, and one of the tools his firm is using to pursue this goal is Salesforce's Agentforce platform. Aum Lifetech is a Philadelphia-based bio-tech company. Its mission is to advance bio-medical research and drug development through innovative genetic research products, specifically gene therapy with a focus on RNA silencing. It has global customers in about 38 countries, including universities and pharmaceutical companies supporting research applications in genetically-defined diseases, including immunology, cancer and various diseases of the nervous system. As part of its day-to-day operations, the organization attends a good many scientific conferences each year, events that provide many of the sales leads for the firm. That's a good source of business opportunities, but it can be overwhelming, Aishwarya explains: We collect leads from these conferences...Once we attended a conference and collectively took 2000 leads go back to our offices. Being a small company with limited resources and limited team members, it's a bit challenging to reach out to everybody at the same time. Otherwise a few people can just email and it may take multiple weeks to reach out to everybody. We always needed something which is more automated. That something is Salesforce Agentforce, which caused Aishwarya to wonder if agentic tech could be used to the firm's advantage: We have been using Salesforce since the beginning of the company. In the CRM, we have all the data for customers already, so it was an easy extension...AgentForce has enabled us to scale outreach to thousands of leads weekly, monthly and turning cold contacts into personalized engagements over months which is essential for a high-value consultative sales model in bio-medical research. We can also do faster client service turnaround so we can answer the questions of clients much faster. Automated meeting summaries, tailored follow ups also allows us to compete with industry giants despite our limited resources. Agentic tech is also enabling productivity gains around developing new systems, adds Aishwarya: I'm a cancer researcher by training. I work at University of Pennsylvania School of Medicine. So I'm not a coder per se...if we can ask an agent to do those things for us, we don't have to wait for programmers...If you have an idea as a biotech CEO you can ask an agent to do this for you, provided you have the governance and knowledge in place. That's something which I think is the biggest power for non-coding CEOs who don't come from a non-coding background. So it gives that immense power to people like me who can now execute their vision without waiting for a big implementation by a 20 people team, which is of course good. That compliance and governance aspect is very important to organizations operating in a highly-regulated space. Aishwarya explains: We work with a lot of biotech, pharmaceutical companies, big companies. We have to be compliant. We deal with a lot of patient data also....This is where Salesforce Data Cloud comes in, where we can ground all our responses to our existing knowledge so that we do not say anything which is incorrect, because in scientific field, scientific credibility is extremely important. For me, knowledge and governance and putting proper guardrails in all of my agents is also very important. Trust is a factor cited by organizations across all sectors as a factor in making agentic adoption decisions, of course. Aishwarya argues that this is "a solvable problem", noting: You have to do the risk analysis as well. I mean, at this point of time, there are some risks, but I think as we are improving the technology every day, every week, this, in the next few weeks or months or maybe a year, this is a solvable problem. At this week's Dreamforce conference, trust has been ne of the most frequently mentioned terms, along with the idea of Digital Labor, with humans and AI agents working together to transform the workforces of tomorrow. That's something that Aum Biotech has factored into its thinking, says Aishwarya: Agents are allowing us also for the humans like myself and my team to do much bigger tasks and the repetitive tasks can be outsourced to these agents. The first one is of course reaching out to thousands of people in parallel and then doing a Q&A also. For example, we made a very nice web app which is linked to our Salesforce where somebody comes to a website and we can do Q&A. If somebody has questions on the product and the protocol, it can be answered. If somebody wants to order something, we can provide the information. Again, the most important thing is when you are working with biotech scientists and with any company which are selling complicated products, the information is extremely important....For us, credibility is extremely important. We cannot make even a single mistake because then the credibility goes away. He adds: Acquiring a customer is the biggest thing for us, so if I can free up my people not to email manually they can do more innovative tasks. They can also target high-value clients as well. They can do tasks which are not easily be performed by the agents, like we can go to more complicated meetings. We have to do experimental design and in some cases we have to do a lot of high level consultation which cannot be performed by agents at this time. Keeping the human touch involved is crucial, he argues: People still crave for human interaction. I'm not saying that AI is going to replace humans because now we can spend that time in creating more human contacts where they need it. If you have 10,000, 200,000 leads, the agents can try to convert from cold to warm to hot and in that fashion and the scientific team can spend more time innovating. This is just the beginning of AI and agentic AI, Aishwarya concludes, urging other organizations to engage with the potential of the emerging tech: Don't be scared. Take a deep breath and just engage yourself into learning more. Try to empower your people with AI. There will be two types of people - people who use AI versus people who don't use AI. Of course we talk about [impact on] jobs and this and that, but I think human jobs will be there because what will happen is that more repetitive tasks - I don't like to use the word low level, but more tasks which are more mundane and can be automated - will go to these agents. But then humans will do bigger tasks, he says, and that's incredibly important: We still have to discover so many things. We still don't have a cure for cancer. We have to figure out space travel. I wish I could go to Paris in five minutes instead of six hours. We have still so many problems to solve. So if humans can spend their time in figuring out those bigger problems...We have to do a lot of things, so don't worry about jobs getting lost because there are still a lot of things to do for humans. We can just utilize our human minds in finding solutions for this bigger problem.

[21]

Salesforce makes its case at Dreamforce for how Agentforce 360 can bridge the 'Agentic Divide' - SiliconANGLE

Salesforce makes its case at Dreamforce for how Agentforce 360 can bridge the 'Agentic Divide' Can AI move from pilots to production at scale in the enterprise? This was the central question surrounding Salesforce Inc.'s Dreamforce conference in San Francisco as the customer relationship management powerhouse sought to make it clear this week that its major bet on agentic AI would do exactly that. At a time when research showing fragmented strategies, weak governance and questionable ROI are limiting AI adoption in the enterprise, Salesforce CEO Marc Benioff (pictured) devoted much of his keynote address today and a follow-up briefing for the media discussion how his firm's latest releases will bridge what he called the "Agentic Divide." It all comes back to data and proper access to it, according to Benioff. "It's a bifurcation," Benioff told the Dreamforce gathering. "Your customers are elevated, but what about your enterprise? If you don't have your data right, you're not going to get your AI right." Salesforce's announcement this week of Agentforce 360 is clearly designed to streamline the path to enterprise implementation and presumed results across sales, marketing, operations, and service. It includes an integrated platform connecting AI agents, Data 360, Customer 360 apps and a conversational interface for humans and agents to work together using Slack. Much of this week's releases have been implemented within Salesforce itself, according to Benioff, who noted that 1.8 million conversations on Salesforce Help had already been powered by Agentforce and an automated response tool was calling back 50,000 customers per week. "We realized at Salesforce that we have to be ground zero," Benioff said. "Have we been perfect in rolling it out? No. Have we been listening to you? Yes." The integration of Slack with Agentforce is a key element in the company's strategy to bring agents closer to where people work in the enterprise. The update released on Monday included a Slackbot for personalized support and Enterprise Search to retrieve company information using a natural language interface. It's an important step in giving autonomous agents more than just an ability to execute tasks. The road to ROI will depend on AI's ability to meaningfully understand the business and provide correct context, drawing on Slack's ability to parse conversational meaning. "Slack has become agentic OS," Benioff said during his keynote today. "You want to be able to achieve not just prompt engineering, but context engineering. AI alone is not enough. It needs all of these capabilities to connect it, to give it context." Salesforce provided plenty of customers to tell their stories of AI progress. An executive from Williams-Sonoma described how the firm built and implemented an agent on its website for one of its high-profile brands. In a briefing for the media following the keynote, Athina Kanioura, chief strategy and transformation officer for PepsiCo Inc. described how her company had rolled out Agentforce to 1.5 million stores globally, with plans to be "agent first" by the end of 2026. Yet even Salesforce's own customers were surprised that enterprise deployment of AI agents had been slower than expected. "I'm shocked that more people aren't adopting all of these things because it helps you serve your customer faster," Williams-Sonoma CEO Laura Alber said during an appearance with the media. In the media session following his keynote, Benioff addressed questions about the slow pace of deployment. He attributed it to the rapid pace of innovation in the AI field, a process that is often complicated and potentially daunting for firms seeking to work their way through changes in core platforms. "This is the moment where technology innovation is outstripping customer adoption," Benioff told the assembled media. "This is architectural changes. This is running their businesses. This isn't just deploying ChatGPT on my phone." Dell Technologies Inc. founder, Chairman and CEO Michael Dell shared a similar perspective in a brief conversation with Benioff during the morning keynote. His company has been adapting to the rapid pace of change as well, with major changes across its product line to reflect the new AI era. Yet, he has also seen other businesses lag behind in AI adoption. "I saw it as machines thinking for us," Dell told Benioff. "If we don't get into this, a new company is going to come along and it's going to put us out of business. We've totally reset the business, reimagined the business. I don't see as many yet at that stage." In addition to questions about Salesforce's business focus on AI, Benioff was also asked during his meeting with the media to address published remarks over the weekend that he thought National Guard troops should be deployed in San Francisco. The company's CEO chose not to elaborate. "I'm going to limit my comments today to Agentforce and Dreamforce," Benioff said. "We have brought in over 200 of our own police to make sure everyone here has a good experience. We're doing everything we can to make this a safe experience. This is San Francisco's largest conference by a factor of 2x." Benioff's focus on Agentforce this week marked the one-year anniversary of the technology's deployment. Since then, Salesforce has released multiple enhancements, including some major updates within weeks after the previous deployment. It reflected how, like its own customers, Salesforce has moved rapidly to keep pace with developments in the AI field. "A year ago, Agentforce was a product; now it's a platform," Benioff said. "It's the fastest growing product in our history. I've never seen anything grow faster. Because of that we've had to go faster."

[22]

Economic Times

Salesforce boosts AI capabilities with global launch of Agentforce 360

Agentforce 360 has 12,000 customers, including social media company Reddit, restaurant reservation platform OpenTable and employment services firm Adecco, Salesforce said. Salesforce said on Monday its artificial intelligence platform "Agentforce 360" would be available globally across its suite of cloud-based tools to help clients automate routine tasks. Software firms are racing to enhance their products with advanced features as competition intensifies in an industry where clients are increasingly turning to AI agents, which can take actions on behalf of users, to help rein in costs. "Agentforce 360 connects humans, agents and data on one trusted platform," Salesforce CEO Marc Benioff said. Agentforce 360 has 12,000 customers, including social media company Reddit, restaurant reservation platform OpenTable and employment services firm Adecco, the cloud software provider said. Salesforce also said its workplace messaging tool Slack would allow employees to use conversational AI in chats to gather data and complete tasks, with enterprise controls and security guardrails.

[23]

PYMNTS

Salesforce Collaborates on AI Projects With OpenAI and Stripe | PYMNTS.com

The company announced these collaborations in press releases issued Tuesday (Oct. 14). The partnership with OpenAI will allow Salesforce's customer relationship management platform to surface directly in OpenAI's ChatGPT chatbot so that companies can query their sales records, review their conversations with customers and build Tableau visualizations, the companies said in a press release. Companies can also use OpenAI's latest frontier models to build AI agents and prompts within the Salesforce platform, according to the release. New integrations announced Tuesday include Agentforce 360 apps in ChatGPT, Instant Checkout and Agentforce Commerce, ChatGPT in Slack, Codex in Slack, and OpenAI in Agentforce 360 platform, per the release. "As consumers, we already get instant recommendations or insights from ChatGPT," Salesforce Chair and CEO Marc Benioff said in the release. "Now enterprises can deliver that same intelligence and immediacy." "Through our collaboration with Stripe on the ACP, we are delivering the unified system designed for the future of agentic commerce, creating a dramatically faster and more personalized path to purchase," Nitin Mangtani, general manager of Commerce Cloud and Retail at Salesforce, said in the release. Salesforce said in April that enterprises' interest in generative and agentic AI was fueling "massive" growth in its data cloud platform. The company has also been expanding its AI capabilities through acquisitions. It acquired Regrello to accelerate its efforts in agentic process automation, Informatica to enhance the agentic AI features of its CRM platform, and Convergence.ai to accelerate the development of its next-generation AI agents.

[24]

Dreamforce 25 - Certinia sets out to meet customers where they are to improve success; the customers respond

A year ago, at Dreamforce 2024 Certinia was just starting out on its agentic AI journey. I caught up with the company's CEO DJ Paoni at Dreamforce 2025, where Certinia as usual is hosting a packed few days of sessions and networking events as one of Salesforce's longest-standing partners. Paoni is clearly very proud that Certinia has now built 44 agents and is beginning to see early adoption for Resource Management and Customer Success agents. At the Certinia One Journey, One Team event, Simon Creasey, VP Customer Success & Lifetime Value, Worldwide for HPE, and Jocelyn Zanasi SVP, Customer Success from Salesforce EMEA, took the time to share their experiences of using Certinia. Creasey's firm went live with Certinia CS Cloud recently because: We wanted to invest in a new Customer Success platform as in the old existing system, our success plans were missing or were not as elaborate as I would like. The customer success teams in the field that were working with the professional services and financial teams were doing it in PowerPoint and Excel as the Customer Success platform had become a silo. So, we decided not to renew and to select another platform. As HPE had already made a heavy investment in Salesforce, it made sense to create a common platform that could view case data natively and so we chose Certinia. We started with evaluating our methodology and process because a good tool won't fix a bad process and so we got all the artifacts and playbooks in place. It then took us just five months to get 600 users up and running using Certinia PS. We identified key Customer Success Managers to become the champions of change management for the new system. These leaders went through the plans with their staff, and adoption was smooth because users could see the system would make them more productive." As a result, Creasey can see improvements in the quality of customer success plans because the system enables staff to track what is important to the customer. The system also means that he can do business reviews with the CEO in Certinia, presenting live dashboards. Having implemented Certinia in its as-a-Service business, HPE is now thinking about how to apply it to the traditional product business. Zanasi, as might be expected of a Salesforce executive, is more advanced in contemplating the role of AI in customer success. She wanted to use Certinia to support the evolution of Salesforce's business from a subscription model to a consumption model for customers. She explains: We need more information to help staff and customers pivot into the consumption model. We need to create a customer health score, to know what product usage looks like and what product debt looks like. We need telemetry and data to share with consumption-based customers, to create a brilliant feedback loop. We are tracking signals from log-in forward. We are monitoring the accreditation level of users, how engaged they are, are they attending events they are invited to. We have to know the health and satisfaction of hundreds of customers and look at deviations in real time. With the new agentic enterprise, customers have to trust us sufficiently to come with us on this journey. Customer success management is how we build and maintain foundational trust with customers, because we can see problems coming and tell them about it in advance. In this way we create the transparency and trust necessary for a lifetime engagement." The KPIs Zanasi uses are considered in terms of lead measures such as touchpoints around a release and lag measures around customer retention. She is excited about the impact of agentic AI and is "piloting everything with Agentforce." One thing she is exploring is the ability to create a small language model (SLM) for each and every customer using data from the Certinia and Salesforce platforms. She is then overlaying an agent onto the SLM so that Slack bot can be asked about each customer. In this way she is she says, "removing time vampires." One thing she is very clear about, however, is that: These problems do not have technical solutions, the challenge is for teams to be adaptive and that needs change management. Paoni sees SI partners as an important element for the company's growth, explaining that: We have done a lot with our partner academy to get the right enablement providing roadmaps of what agents can do. A partner such as Diabsolut which implements internally, takes the journey and then becomes the evangelist is how we like to work. Naeem Khalid, VP Customer Success, Diabsolut, agrees, adding: We are part of Certinia's early adoption program for new capability. We have not launched agents internally yet and so, similarly, with customers we have no go-live yet. We use PS and CS Cloud ourselves and most of our work with customers is integrating PS and CS cloud to enable score health tracking for customers and projects, making data more accessible. We don't have to build custom code to enable this. Because we use the product, we can share our experience with clients and Certinia spends time with us to listen to our feedback. They share their roadmap, so that if something is not on it, we can go and build a feature. While the spotlight at Dreamforce 2025 is firmly on Agentforce 360 as an exciting development, the solid growth and development for many continues more quietly on core existing functionality. Certinia is a good example of a vendor capably enabling both early adopters for agentic AI and more mainstream customers.

[25]

Salesforce introduces new OpenAI, Anthropic integrations - SiliconANGLE

Salesforce Inc. is expanding its partnerships with OpenAI and Anthropic PBC to automate repetitive business tasks for customers. The new collaborations, which were announced today at Dreamforce 2025, focus on Agentforce 360. It's a suite of artificial intelligence features that Salesforce ships as part of its cloud services. Agentforce 360 includes more than a half dozen automation tools that speed up tasks such as creating ad campaigns. It also enables companies to develop custom AI agents. New Agentforce 360 integrations for ChatGPT will enable users to access Salesforce data via the chatbot service's interface. A contact center manager, for example, could request a summary of support requests from the past week. ChatGPT can also visualize the data that users retrieve from Salesforce in Tableau-powered dashboards. "As consumers, we already get instant recommendations or insights from ChatGPT," said Salesforce Chief Executive Officer Marc Benioff. "Now enterprises can deliver that same intelligence and immediacy." GPT-5, the suite of AI models that powers ChatGPT, will become available in Agentforce 360. The integration will enable companies to incorporate OpenAI's models into the AI agents they run on Salesforce. The algorithms are also set to become available in several of Agentforce 360's pre-packaged AI apps, starting with the Agentforce Sales sales automation tool. Salesforce's collaboration with Anthropic has a similar focus. The cloud giant will integrate an Agentforce 360 automation app called Agentforce Financial Services with Claude for Financial Services, a version of Anthropic's chatbot geared toward banking professionals. Salesforce says that the integration will make AI agents deployed on its platform better at tasks such as investment portfolio analysis. Anthropic's large language models will be available in Agentforce 360 alongside GPT-5. According to Salesforce, the goal of the integration is to help organizations in regulated industries build AI-powered automation workflows. Highly regulated companies often have stricter cybersecurity requirements than other enterprises. To address those requirements, Salesforce will keep all the traffic generated by Claude inside its virtual private cloud. Anthropic is the first LLM provider whose traffic will be processed in that manner. "Regulated industries need frontier AI capabilities, but they also need the appropriate safeguards before they can deploy in sensitive systems," said Anthropic co-founder and CEO Dario Amodei. "We've built Claude to deliver both: the performance and the safeguards." Both Anthropic and OpenAI will roll out new integrations for Slack as part of their respective partnerships with Salesforce. The connectors will enable Slack users to retrieve data from Salesforce services with natural language prompts. The companies also plan to roll out other features, including one that will make it possible to give OpenAI's Codex programming agent access to information in Slack channels.

[26]

Dreamforce 25 - Agent Fabric is déjà-vu all over again for MuleSoft

Every enterprise today can select from a rapidly expanding choice of AI agents on offer from all of the leading technology vendors. But behind all the hype about what these digital automatons can achieve, there lurks a growing danger that their rapid spread will prove unmanageable. It's a quandary that feels remarkably similar to the challenges enterprises previously faced as they moved to the cloud, says Andrew Comstock, SVP and GM of integration vendor MuleSoft: I think about what happened with digital transformation. You look at, over the last decade or 15 years, so many companies leaned in on, 'We're going to modernize our systems. We're going to modernize our businesses by taking advantage of cloud technologies' -- whether it's SaaS or these hyperscalers. [They did it] really successfully. But it did something unexpected that people hadn't planned, which was it caused a bunch of explosions of complexities, whether this was teams building new APIs and new microservices, whether this was buying too many applications and not figuring out how to connect them together. MuleSoft built its entire business on the back of that complexity, where we helped companies solve for those use cases. That's what MuleSoft has been about for so many years, is enabling people to operate and connect all of those systems together. Well, what we're starting to see is in the AI space, a lot of the same hallmarks. Companies are increasing their adoption of AI and agents. This is a new kind of fragmentation, this agent fragmentation, agent explosion. So we're seeing connected, chaotic, disconnected workflows, which is hitting on the same challenges we saw in the past, and for us, this feels a lot like déjà-vu. In response, Salesforce-owned MuleSoft recently introduced Agent Fabric, a management platform where enterprises can register, orchestrate, govern and observe every agent in their IT landscape, whatever its origin. There are four components, all of which are slated for general availability this month: Comstock says that customers have started to ask for these capabilities as they recognize a growing need to manage governance, compliance and security across an increasingly complex agent landscape. He says: One of the questions that I think we're hearing [from] customers is, they're going to start shifting from, 'How do I build my agent?' to, 'How do I enable these agents, no matter where they're built, to start working together more securely, more autonomously, and more with better purpose?' That's what the Agent Fabric is designed for. It's bringing that feeling of déjà-vu and so many of the things we've done in the app world into the agent world. So this is enabling customers to control that explosion, by orchestrating and governing every single agent across your agentic enterprise, no matter where it's built... When we talk to people about it, they go, 'Oh, this is what I bought you for. You're getting back to the heart of what we loved you for.' This is us bringing that love back to the next generation. This is a field where the technologies continue to evolve, and therefore there's still no established mechanisms for handling them, and industry standards are still emerging. While protocols such as Model Context Protocol (MCP) and Agent-to-Agent (A2A) have been introduced, they lack many of the components that enterprises rely on to keep data and systems safe and secure, such as robust authentication and access permissions. Veterans of the mashups era, when APIs and web applications first started to be used in the mid-2000s, have been likening today's MCP integrations to the 'Wild West' integrations of those early days. Comstock agrees that there's still work to be done to iron out the rough edges: This is where the technology is, at the moment, still the limiting factor. With capabilities like Agent Fabric and some of these other identity providers scaling up in their agentic support, I think you'll start seeing the proliferation at scale extend. Again, the parallels to me for cloud, I think, are very high, because we saw the same challenges with cloud, which is, 'How do I manage who has access to these services?' It took the vendors a little bit to catch up, and we're just seeing those services start coming together. Similarly, there are gaps today in observability that need to be filled. He explains: "We've built into place ways that we can start understanding what agents are talking to what services and what agents are talking to other agents, and helping you visualize that connectivity tissue. So at least you can start identifying the services. " Down the road, I think you'll see more of the interlocking of the actual tracing across agent stacks to make that more productive. That's, I think, something that will evolve out. Again, this is where more data is going to need to be available from the agent providers, for people to really start tracing through the decision-making that some of these agents are doing. All of this is a reflection of the lack of established practices that are well understood. He goes on: One of, I think, the most resounding truths in the agentic landscape right now is there isn't a best practice. You can go out today buy 'Cloud for Dummies' and buy a book that tells you exactly how to build these things. There is no 'Agent Development for Dummies' yet. In a decade, there will be. And so as part of this, we're seeing companies try different types of patterns and different types of architectures. Back in the day, before this shared knowledge about cloud became commonplace, many enterprises established centers of enablement staffed by API experts, who were able to advise colleagues on how to best deploy cloud services and avoid inadvertently creating vulnerabilities or excessive technical debt. Those centers have now largely been disbanded, but Comstock sees them making a comeback in the era of AI agents: It's almost funny, like we forgot why we did it. We did it because a decade ago it was hard. Fifteen years ago it was hard. We built it because we wanted to try to force behaviors to happen. I've seen some of the best companies now start doing the same thing with AI, where they're bringing inside, like the CIO or the CTO's organization, they're spinning up a Head for AI. Those organizations are increasingly looking like the centers of excellence that we knew from a decade ago, focusing on cloud, focusing now on AI, where they're pushing out messaging, they're pushing out best practices to enable and empower their teams. I think you'll see some of that come back. This is again why I keep coming back to déjà-vu. We know what to do. We know how to go through this transformation. It hasn't been so long we've forgotten. Let's bring out those same techniques to accelerate the cycles. Adding enterprise-grade governance and security as agents become a fixture in the IT landscape is as pressing a concern today as it was in the early days of REST and other forms of lightweight API integration two decades ago, as cloud services began to proliferate. While the technology is different and the pace of innovation and adoption is far faster today, we can still learn a lot from that history. If anything, it's even more critical to put management processes in place today, because autonomous agents are capable of exploiting vulnerabilites and compounding errors at a far greater speed and scale than earlier automation techologies. The quandary for enterprises is choosing which management platform to adopt, as many application vendors seem to want to also become IT vendors and offer their own agent management capabilities alongside their own agents. As well as these incumbent application vendors, most enterprises also have longstanding relationships with integration vendors and may well feel more comfortable with a platform that can be a neutral player between the various application platforms within the IT estate. In this regard, MuleSoft is something of a hybrid, being owned by Salesforce, but still relatively independent in its commitment to integrate across multiple platforms. It will be important to continue emphasizing that neutrality as it advances its offerings for the agentic enterprise. Check out diginomica's dedicated Dreamforce event hub here.

[27]

Salesforce turns Slack into a platform for conversational AI agents - SiliconANGLE

Salesforce turns Slack into a platform for conversational AI agents Salesforce Inc. today unveiled major updates to its Slack platform that brings artificial intelligence agents closer to where people work by making the platform into a unified conversational interface. Today's update includes advances that surface core Salesforce apps directly in Slack using AI, such as Tableau Next, Agentforce Sales and IT and HR Service, combining customer relationship management with conversation. Two major upgrades to Slack introduce an all-new Slackbot for personalized support and Enterprise Search, which uses natural language to retrieve company knowledge. "By making Slack the conversational interface for Salesforce, we're giving every employee a trusted, unified home for AI and agents -- and transforming how work gets done," said Slack Chief Executive Denise Dresser. Part of the newly announced Agentforce 360, AI agents capable of accessing CRM data directly are now embedded in Slack, connecting users to Salesforce. Employees can "talk" to their data and platforms just like other teammates and receive conversational or structured information, or ask AI agents to take action on their behalf. A new native Agentforce agent in Slack, called Channel Expert, is also rolling out. It delivers always-on expertise within any conversation, using real-time search and user-provided content to give up-to-date answers to common questions. When it can't answer, it flags teammates who can jump in to close the loop. Slackbot has been reimagined as a personalized AI companion for teams that can access the breadth of Slack conversations, files and enterprise context to tailor its responses to match team cadence. Over time, Slackbot will learn user preferences and work styles, using that information it will be able to offer information and insights tailored to how people work. The company said this will allow Slackbot to provide support that feels tailored to personal needs, summarize threads to pull out key information that users often look for and deliver more natural feeling assistance for drafting updates and organizing projects. "Slackbot is showing us what the next era of productivity looks like: AI that works alongside every employee as a personal companion," said Andy White, senior vice president at Salesforce for Salesforce technology. "For Salesforce, that means our teams spend less effort piecing work together and more time delivering value for our customers." Now accessible in Slack via a search bar, Enterprise Search connects to all knowledge across a company using natural language questions. Salesforce announced upcoming connectors to third-party services including for Google LLC's Gmail, Microsoft Corp.'s Outlook, Dropbox Inc. and Notion Labs Inc. -- with the addition of custom connectors to internal systems -- users can securely access company knowledge with a single search. Because Slack is where teams collaborate and bring together the language of work, Salesforce said, it provides a wealth of data for AI agents to work with and the richest context comes from conversations. To support third parties building on this customer-owned and controlled conversational data, Salesforce created standardized connectors, including a new real-time search application programming interface and Model Context Protocol Server. "[We] realized Slack is the perfect AI agents platform," said Vercel Inc. CEO Guillermo Rauch.. Slack's platform gives us the perfect foundation, and our agents don't feel like tools -- they feel like team members." A number of companies are already building on these new capabilities, including Vercel, Google, OpenAI, Anthropic PBC, Perplexity AI Inc. and Cursor. With access to these APIs and tools, they can securely tap into Slack's conversational data to build more intelligent agents that connect to and understand team collaboration.

[28]

Dreamforce 25 - how PwC is using Agentforce to help customers re-think customer engagement

As Dreamforce 2025 swings into action, PwC took the opportunity to launch its Agentic AI-Powered contact center offering. It makes use of Salesforce's Agentforce Service to integrate conversational AI, predictive ordering and unified customer data to create AI-first interactions. As I was in San Francisco for Dreamforce, I popped into PwC's offices there to speak with Patrick Pugh, the firm's Global Alliance & Ecosystems Leader and Derek Santana, PwC's US Salesforce Alliance Leader to find out more about the offering and PwC's partnership with Salesforce. PwC has had a strategic partnership with Salesforce for over a decade and saw the potential impact of working with Salesforce Agentforce around contact centre transformation. Working with Salesforce Product Leadership and the Agentforce Service team, PwC created its offering. Pugh explains; The contact center space is ripe for transformation using AI, because it is a high-cost operation that is fragmented and often generates low levels of customer satisfaction. For this reason, it was a logical area for Salesforce and PwC to lean into together. The partners worked together to develop the offering for a multinational consumer goods company with PwC acting as orchestrator, a term the firm prefers to systems integration. The metaphor was chosen Pugh says to convey the idea that PwC is about more than providing technology, as: We understand different ways of working with the process, we understand the industry and, as well as depth of technical understanding, we also know how to help customers change their ways of working. We make all these pieces come together, using Salesforce and other technologies that clients have in their IT estate. We help them harness and improve their data as well as understanding the current technology capabilities and their future direction. Santana continues: We help clients see through the fog get the outcomes they want because this is never a single technology solution, we have to work with the technology cobweb that is in place and work out how to apply Agentforce. He adds that: The multi-national client was experimenting with AI but wanted help realising the technology's potential at scale and to work with a partner that could provide a capability that would be secure, trustworthy and reliable. Salesforce wanted to go to market in an intentional way with a partner that could provide clients with strong cybersecurity and compliance expertise. In its press release for its Agentic AI-Powered Contact Center offering, PwC states that its AI-first model could provide a projected 70-80% cost reduction over five years through Agentic AI workflows coupled with AI augmentation for the workforce. The business case for the first client is based over the five-year contract as PwC makes the contact centre more efficient. According to Pugh and Santana this will happen as the client works with better, cleaner data sets, and frontline personnel develop capabilities they currently don't have via training and changing hiring practices. Santana explains: For the frontline personnel we are developing a range of persona types, because an offering with a uniform agentic response will not be transformational. The client has bought into the transformation across the company and understands the importance of workforce enablement to elevate the skills base. As Pugh comments: With a client of this size, moving the needle a point or two at scale makes a massive difference. PwC is also predicting a 10-15% enhancement in CSAT scores for clients that deploy its Agentic AI-powered contact centre offering. When asked how, Santana's answer is, "speed to resolution for the customer." I wondered if the improvement might depend on whether CSAT scores were currently below industry average, but Santana believes that there will be a positive uplift regardless. The deployment is still in flight, and so it is too early to see the realisation of these predictions for the first client, but PwC is confident that they can be delivered, not least because there is an outcomes-based element to the contract. A year on from the launch of Salesforce's Agentforce Partner Network, Pugh says that: People are moving beyond the pilots and are looking to get the most out of the technology. They have a thirst for outcomes and AI is part of our reality for delivering that. Santana explains that there is a big opportunity to rethink business models, to expand from B2B to B2C and vice versa, to consider ways to disrupt existing ways of doing business. He concludes: Agentic AI is not a concept anymore as we are beginning to get production results. The IT services sector has been pursuing the holy grail of outcome-based results for decades now. It is important because its successful enablement points to an industry that can use technology to actually align service delivery with business requirements. There is a lot of excitement about the impact agentic AI may have on realising outcome-based results. Pugh told me that: AI is raising expectations around delivering payment on outcomes. If it can deliver it would be a market gamechanger. For example, it would provide the reason to believe that when a company such as PwC publicly states percentage improvements provided by an offering, they are probably true. Why? Because they would be literally staking their fees on their predicted results. Check out diginomica's Dreamforce event hub here.

[29]

Salesforce bets big on agents with platform overhaul - SiliconANGLE

Salesforce Inc. is going all-in on artificial intelligence agents with this week's announcement of Agentforce 360, an enterprise-wide platform that infuses agents into nearly every application the company delivers. Salesforce calls the platform an attempt to reimagine how AI and humans work together. "It's probably the biggest transition in technology I've ever experienced in my career," said Parker Harris, Salesforce's co-founder and chief technology officer. Agentforce 360 is a bet that every enterprise will want to have a fleet of AI agents working with humans. Agents will reason, take actions and adapt to a company's culture, voice and business logic. The platform introduces a new generation of AI agents with hybrid reasoning, voice interaction and deep hooks into Salesforce's data and workflows. The company positions it as a full-stack framework for building agents that can not just answer questions but get work done. At the heart of the platform is Agent Script, a JavaScript Object Notation-based language that enterprises can use to define complex, deterministic workflows, such as following refund policies or qualifying leads. The release also unlocks new capabilities in the Atlas Reasoning Engine, the Agentforce component that enables agents to understand human intent, plan and autonomously execute tasks. Atla is described as a hybrid system that combines the probabilistic thinking of LLMs with the precision of business rules. That means it can interpret requests phrased differently while still behaving predictably. "Imagine asking your bank for your balance and getting a different answer each time," Harris said. "That won't work. You need structure, memory, and governance." Voice is a critical element of the new release. Salesforce's "eVerse" framework promises low-latency, high-fidelity and relevant voice experiences that eliminate customers' frustration at waiting on hold or hunting for a key that summons a human operator. Slack, the collaboration platform that Salesforce acquired nearly five years ago, will be the principal interface for the agentic experience the company envisions. "Slack is becoming your agentic operating system," Harris said. Slack is where agents will talk to each other, escalate issues to humans and pull in data from around the company. Slack is also getting some new AI features, including natural language search and a new agent called Slackbot. The company plans to support Microsoft Corp.'s Teams, but executives made it clear that Slack is the preferred interface. "We want Slack to be like water; just part of the Salesforce experience," Harris said. To make it easier for customers to build their own agents, Salesforce is also rolling out Agentforce Builder, a conversational development environment with a document-style canvas, real-time simulations and support for "vibe coding," a popular new software development approach in which a developer describes a project to a large language model and it generates the code. The company says it will protect against some of the security and quality issues with vibe coding by ensuring that agents adhere to the same standards as those defined by Salesforce. "Every agent is versioned, testable, portable and governed out of the box," said Srinivas Tallapragada, Salesforce's president of engineering and customer success. The platform is capable of turning unstructured documents like contracts and catalogs into structured, queryable records. It's Salesforce's way to fix what Tallapragada calls the "prompt doom loop, where you're writing prompts and getting frustrated because the context is not there." Salesforce will use context indexing and a feature it calls Data 360 to short-circuit that loop by grounding agents in governed, up-to-date, role-specific knowledge. Data 360 unifies all of a company's structured and unstructured data and metadata, while context indexing automatically extracts and structures business-specific information so agents can deliver relevant outcomes. Salesforce said the index inherits the same security, privacy and compliance controls as the rest of an enterprise's data. The company said it has deployed Agentforce across its own service teams and now uses agents to handle over 1.8 million conversations each week. It said the results have been fewer repetitive tasks, more proactive outreach and up to a 40% increase in proactive service activity. The sales organization uses AI-driven representatives to follow up on thousands of long-tail leads that previously fell through the cracks. "That's what we mean by an agentic enterprise," Tallapragada said. "It's not just turning on a tool; it's rethinking how you do sales and deliver service." Heathrow Airport has deployed an agent called "Hally" to field real-time requests ranging from wait times at security to automatically extending parking limits. "It pulls in complex back-end data into a simple, human-like interface," said Heathrow's Peter Burns, Heathrow's director of marketing and digital. "It lets us scale personalized experiences." Williams-Sonoma Inc. is using interior design agents to help customers avoid decorating mistakes. In the process, average order value has increased. "This isn't just about automation," said Williams-Sonoma CTO Sameer Hassan. "It's about augmenting creativity. When you equip brilliant people with this kind of tech, you get better outcomes, strategically, creatively and financially."

[30]

Salesforce strengthens its alliances with OpenAI and Anthropic and integrates Agentforce into ChatGPT

On Tuesday Salesforce announced a major expansion of its partnerships with OpenAI and Anthropic, along with the integration of Agentforce 360, its customer management platform, with ChatGPT by the end of the year. This connection will enable companies to directly access their Salesforce data from OpenAI's assistant, generate visualizations in Tableau, and, from 2025, sell products using ChatGPT's built-in instant payment feature. At the same time, Salesforce is preparing a similar integration with Claude, Anthropic's model. Brian Landsman, CEO of the AppExchange division, presented this development as a structural change in the way users will interact with software, whether via ChatGPT or Slack. The announcement comes on the eve of the Dreamforce conference, at a time when the group is seeking to revitalize its growth after a 26% decline in its share price since January, while the S&P 500 has risen 13%. On Monday, Salesforce already unveiled the addition of a voice feature to Agentforce, confirming its strategy of enriching the user experience. These partnerships are part of increased competition among software publishers to integrate with major generative AI models. After Atlassian, Datadog, and Intuit, Salesforce is betting on OpenAI and Anthropic to strengthen its presence in regulated sectors, particularly financial services. The group also plans to adopt Claude Code, Anthropic's programming tool. Marc Benioff, CEO and co-founder, emphasized that these partnerships are based on a close technological relationship: "All of these next-generation AI companies, from OpenAI to Anthropic, use Slack as their operating system to run their businesses."

[31]

Dreamforce 25 - another day, another OpenAI partnership announcement, but this one is "the first of a kind" apparently

A couple of years ago OpenAI CEO Sam Altman caused a bit of a stir at Dreamforce when he tried to convince the audience that they should regard hallucinations as a beneficial side effect of AI. This brave/ludicrous -delete as applicable - stab was made to Salesforce CEO Marc Benioff's stoney face, coming as it did only a few hours after he had railed against the self-same hallucinations in his opening conference keynote: I call them lies! That wasn't the only controversial utterance from Altman that would have raised his Salesforce counterpart's eyebrows - such as his pronouncenent last month that we're entering an era of so-called "fast fashion" SaaS, signaling a shift towards rapid, AI-powered software development and deployment that will disrupt the market status quo. Flash forward to this week's Dreamforce and OpenAI and Salesforce have taken their relationship a significant stage further in what is being pitched as a "first of its kind partnership". According to the official announcement blah blah: [This] will let companies access Salesforce's Agentforce 360 platform in ChatGPT -- querying sales records, reviewing customer conversations, or building Tableau visualizations, simply by typing a query into ChatGPT. They can also use OpenAI's latest frontier models, including GPT-5, to build AI agents and prompts within the Salesforce Platform. And a new commerce experience will give shared customers the ability to sell to hundreds of millions of potential US users in ChatGPT while keeping full control of their processes, data, and customer relationships. Drilling down, the announcement includes: The pitch for the last two of these is basically that having ChatGPT in Slack gives teams the power to summarize complex conversations, or quickly draft posts and replies, while in the flow of work, while Codex in Slack enables teams to tag @Codex in a channel or thread on order to delegate programming tasks, like they would with a co-worker in the room with them. These Slack aspects of the new partnership got particular attention at the opening of Dreamforce with Brad Lightcap, COO at OpenAI, telling delegates: OpenAI started on Slack in many ways. It was the first place that I logged in when I joined OpenAI, and was the fire hose that I was drinking from when I was trying to understand what it meant to build AI and AGI. And I think one of the really special things about Slack is it moves at the pace that the company does. So it's this really rich picture of all the projects we have going on. It brings all of the right people together exactly at the right time, and it allows us to organize and re-organize, which has been the DNA of AI research at OpenAI since the beginning. And so it's the way I think modern companies work. He added: What makes AI special is it allows you to do the work between work. We've got all of these software systems now that define the modern enterprise, and what AI really allows you to do is start to stitch between them in a way that supercharges them. So it brings context from System A, System B and System C together. It makes those insights really actionable. Eventually, it can actually take action for you as AI becomes increasingly capable in using tools. That's really what's going to ultimately supercharge enterprises - the ability for AI to supplement people on top of the systems alongside the systems that already existed. A partnership with OpenAI is practically de rigueur these days in the enterprise AI space, but this is an interesting formalization of the relationship between the two companies. No sign of Altman in public here yet - will he be popping up at Oracle AI World over in Vegas, given the huge amount of money he's paying that firm to build out the infrastructure he needs to support his growth ambitions for OpenAI? - but he and Benioff have been bumping into one another around the industry over the past few months (not least most recently when both were guests of King Charles at the Royal State Dinner staged for President Donald Trump's visit to the UK! It really is who you know, you know...) And no hallucination-defending this time around as Lightcap told the Salesforce audience of 2025: Our vision for the Future is simple. We believe that AI systems are going to unlock potential for humanity levels that we today can't even comprehend. And so how does that happen? One is, we've got to integrate with the systems and tools that people use. Two is, we've got to be able to create new experiences for people to do more. And three is, it's all got to be trust based, and we've got to get that right. That sounds much more on message with the Benioff playbook...

[32]

Salesforce aims to shake up IT service delivery with AI agents - SiliconANGLE

Salesforce aims to shake up IT service delivery with AI agents Salesforce Inc. is imbuing its 16-year-old Service Cloud with artificial intelligence agents in a new suite designed to deliver what executives described as a "conversation-first, agent-first" experience. The company is positioning Agentforce IT Service as an evolution from ticket-based information technology service management to an interactive dialogue with autonomous resolution built into widely used collaboration tools such as its Slack platform and Microsoft Corp.'s Teams. The new cloud application provides an agentic IT service desk with the ability to answer common questions, generate service tickets and even remotely diagnose and repair some problems. All functions are compatible with the Information Technology Infrastructure Library, a global framework of best practices for IT service delivery. Multiple AI agents can work together as subject-matter experts to support employees and IT teams in resolving complex problems. Connectors, integrations and workflows help integrate with an organization's existing IT portfolio. The new product grew out of customer demand for a more employee-centric approach to service, said Kishan Chetan, executive vice president and general manager of Salesforce Service Cloud. "We've consistently heard that our customers want the same experience that we enable them to provide to their customers to their employees," he said. " They want to engage with employees in a personalized, proactive manner." Salesforce cited recent research by WalkMe Inc. that found that the average enterprise wastes more than $1 million a week in lost productivity due to IT support inefficiencies. Chetan said traditional ITSM systems have struggled to meet modern expectations. "Teams have spent a lot of money trying to automate IT service," he said, "but most of the time, questions are translated into a form and a ticket." The outcome, he added, is a slower response and more work for IT teams. "As you create tickets and more issues for the IT team, you slow them down," he said. Salesforce said its Agentforce IT Service reimagines support around a conversational interface and delivery through the applications employees prefer. "Whether they're in Slack, Teams, email or wherever, employees have the ability to get the right information right there," Chetan said. Instead of submitting tickets and waiting for responses, employees can ask questions such as whether they qualify for a laptop upgrade. The system automatically checks individual profiles and company policies to provide tailored answers without IT intervention. The platform also takes a proactive approach to managing some problems. During widespread issues, such as an email server outage, Agentforce can alert affected employees, deliver live status updates and track incidents through to resolution. For complex or sensitive situations, Agentforce can route requests to IT personnel along with relevant context and diagnostic insights. Salesforce said these capabilities together move IT support from a reactive, ticket-based model to a conversational, predictive and collaborative approach. Agentforce IT Service builds on Salesforce's existing Service Cloud, which about 70,000 customers use. Asked if the company intends to leverage its large installed base to challenge category leader ServiceNow Inc., Chetan said the strategy is broader than that. "There's a lot of unfulfilled demand in the market," he said. "I'm sure we'll run into ServiceNow, but we're very confident of our continued innovation." Muddu Sudhakar, Salesforce's senior vice president and general manager of IT and HR service, echoed that sentiment. "I think the world will be much bigger where we'll coexist," he said. With concerns about the trustworthiness of generative and agentic systems topping the agendas of enterprise leaders, Sudhakar said keeping humans in the loop was a core design principle. "It's very important that every step of the way, humans are involved," he said. "Live agent escalations provide guardrails so AI agents don't deviate and become rogue. The employee has the option at any point to go from talking to an agent to a live human IT team member." Omar Baig, chief information officer at the United Nations Educational, Scientific and Cultural Organization, called Salesforce "a mission-critical app platform" and expects to deploy the new service to support more than 60 offices worldwide. "The key challenge that I'm facing is how to deliver more with less," he said. "We need the tools and platforms to [provide support] with zero downtime." Chetan said Salesforce expects agents to spur a large-scale replacement of legacy help test systems. "The industry is going through a rapid transformation with agents and conversation first," he said. "There's a fundamental shift that's going on in the market."

[33]

Salesforce Teams Up With Stripe, OpenAI on Instant Checkout

Salesforce is collaborating with OpenAI and financial-services company Stripe to build a way for its merchants to tap into artificial intelligence-powered shopping with a new instant-checkout tool. The cloud-software company on Tuesday said the partnership to build and integrate the feature into its Agentforce Commerce will allow for faster purchases and new growth for its customers' digital storefronts. With the tool, AI agents will facilitate checkouts, helping merchants capture demand, Salesforce said. The collaboration comes after OpenAI said last month that ChatGPT users could buy things directly on its popular chatbot, with Stripe processing the payments. OpenAI at the time unveiled an open-source technical standard for merchants to build integrations with ChaptGPT, called Agentic Commerce Protocol, which Salesforce is now using to give its merchants instant-checkout capabilities. San Francisco-based Salesforce said that, according to its own research, nearly half of shoppers who already use AI for purchases are open to having an AI agent make a purchase for them.

[34]

Salesforce introduces voice to Agentforce, its AI platform for customer service

On Monday,Salesforce unveiled Agentforce Voice, a new feature that allows its Agentforce artificial intelligence platform to handle phone calls. This major development, announced on the eve of the Dreamforce conference in San Francisco, extends the capabilities of the software already used by over 12,000 companies for text-based interactions. Users will now be able to customize the voice, tone, or speech speed of AI agents and even interrupt the conversation for a more fluid experience. Compatible with Amazon, Five9, Nice, and Vonage phone systems, Agentforce Voice integrates directly with existing call center infrastructures. This launch illustrates the rise of voice in generative AI, following an initial phase focused on text. Salesforce is thus positioning itself against growing competition from players such as OpenAI, Anthropic, and Sierra, the startup founded by former co-CEO Bret Taylor, which is also developing enterprise voice agents. Despite this innovation, Salesforce is going through a period of stockmarket tension: its share price has slumped 28% since January, while the Nasdaq has risen 15%. Investors are concerned about the impact of AI on business software, a risk acknowledged by the company itself. However, CEO Marc Benioff is seeking to reassure investors, believing that the transformation will be gradual. Salesforce also plans to launch Agent Script in November, a new tool for customizing the responses and behaviors of its virtual agents.

[35]