The Rise of AI Agents: Transforming Business Operations and Customer Interactions

New AI Agents Take on Management Jobs | PYMNTS.com

A new breed of autonomous artificial intelligence called agents is taking over customer service and operations at major companies, making decisions that, until recently, required human managers. The shift marks a turning point in enterprise automation as companies move beyond basic chatbots to deploy sophisticated AI agents that can navigate complex business processes independently. Commerce.AI's latest offering exemplifies this trend, with AI systems increasingly trusted to handle sensitive customer interactions and operational decisions that shape company performance. Commerce.AI launched auto-AGENTS, an AI system that autonomously handles business interactions across voice, chat and text channels. The system includes specialized AI agents for tasks like data retrieval, sentiment tracking and post-call automation, aiming to boost enterprise efficiency while allowing human agents to focus on high-value work. "While traditional AI approaches have centered around assistance, the ability for AI agents to reason, decide and take action will amplify results," Archana Kannan, senior vice president of product for work messaging app Slack, told PYMNTS. "Ultimately, agents are going to transform how every user gets their job done, particularly the mundane, common tasks like automating projects, new hire onboarding, generating content or managing IT incidents." AI agents are autonomous software systems that can understand the context and make decisions across multiple business tasks. Unlike basic chatbots, they can handle complex workflows independently, from managing customer support tickets to orchestrating supply chain operations. AI agents have been stealing the spotlight in tech innovation. OpenAI's new Swarm framework launched last month allows developers to orchestrate multiple AI agents for complex tasks, enhancing collaborative problem-solving capabilities. Skyfire Systems introduced a payment network in August tailored for AI agents, enabling autonomous financial transactions and signaling a shift toward self-sufficient AI operations. These developments highlight a transition from isolated AI systems to interconnected agents, poised to revolutionize commerce, logistics, and creative industries. ServiceNow is deploying AI agents to handle IT and customer service issues autonomously. At The Ottawa Hospital, similar technology provides patient information while reducing medical staff paperwork. Commerce.AI said in a Nov. 14 press release that its new auto-AGENTS system can track customer sentiment and retrieve internal documents in real time. Its features are aimed at highly regulated industries like healthcare and finance. The AI software connects with existing enterprise systems like customer relationship management systems and data lakes, although the privately-held company didn't disclose pricing or current customers. "This system doesn't just assist -- it autonomously handles tasks," Commerce.AI CEO Andy Pandharikar said in the release, adding the technology is designed to let human agents focus on more complex work. The system builds on the company's auto-MATE platform, which was introduced last year. Kannan said AI agents will complement -- not replace -- human connections. By using real-time structured and unstructured data to personalize interactions, AI can reflect a company's brand voice while automating routine tasks. "At the same time, employees stay involved for moments that require a personal touch," Kannan said. "Agentforce agents in Slack are a great example of how this balance works. Agents act as an extension of the team, seamlessly stepping in to handle tasks, while humans remain available to build and nurture relationships. The result is agents that enhance, not detract, from customer trust." As AI agents reshape how companies work, experts say we need new ways to measure their impact. Karli Kalpala, head of U.K. and Ireland and strategy transformation at Digital Workforce, told PYMNTS that organizations must rethink how they measure AI's value in business operations. "The brilliance of AI agents is that they are trained to the unique contexts of each business, and as such, demonstrating ROI has to be understood beyond traditional metrics," Kalpala said. Success requires enterprise-grade platforms to ensure business user control and robust security, said Kalpala, who specializes in enterprise automation transformation. Forward-thinking companies use AI to predict customer inquiries and perform real-time fraud detection while balancing automated efficiency and human expertise. "Leading companies aren't choosing between automation and human connection," she said. "They're redefining how these elements work together, combining their strengths to deliver excellent customer experiences." Ankur Sinha, chief technology officer of Remitly, a digital money transfer service that helps immigrants send money to families abroad, said AI agents must prioritize customer trust over operational efficiency as FinTech companies race to deploy the systems. "Efficiency gains may impress shareholders, but trust is what wins customers," he said. "The true ROI of agentic AI lies in its ability to deliver consistent, reliable experiences that build loyalty over time." The digital payments company, which operates in more than 170 countries, is developing AI systems that recognize cultural nuances and adapt to individual customer preferences. However, Sinha stressed that human oversight remains crucial. "Automation and personalization don't have to be at odds," he said. "At scale, personalization can turn customer interactions from transactional to relational, building loyalty and trust." Cognizant Chief Technology Officer of AI Babak Hodjat said business metrics, not technical accuracy, should drive the assessment of autonomous AI systems. "We should think of AI workflows as leading to business decisions that impact business KPI," he said, citing trading systems as an example where model accuracy matters less than actual risk-return performance. Hodjat, who leads Cognizant's AI lab, stressed the need for safeguards, including human oversight triggers and a "disengage button" to halt autonomous operations if needed. The system should then "fall back to an entirely manual or fully predictable mode of operation."

Thousands of AI agents later, who remembers what they do?

Gartner weighs the pros and cons of the latest enterprise hotness Among the optimism and opportunities perceived around AI agents, Gartner has spotted some risks - namely that organizations might create "thousands of bots, but nobody now remembers what those bots do or why they were built." The collision of muddied management thinking and much-hyped autonomous agents will be interesting to watch play out. In Gartner's view, "agentic AI" is about "goal-driven software entities that have been granted rights by the organization to act on its behalf to autonomously make decisions and take action." They differ from robotic process automation, which stitches together enterprise applications with trained bots, as agents do not need explicit inputs and their outputs are not predetermined. The analyst company notes that AI agents have become the flavor of the month with vendors. Notable examples include Salesforce, with the vendor's ebullient CEO, Marc Benioff, boasting to investors that by releasing a billion agents by 2026, Salesforce could capture a "very high margin opportunity." In its latest paper (available only to clients, although there is an upcoming open to all webinar), Gartner weighs up the pros and cons. Perhaps it's wisest to start with the bad news. "The danger exists of repeating the robotic process automation problem: organizations created thousands of bots, but nobody now remembers what those bots do or why they were built," it says. In addition, organizations might also build and deploy their own low-code agentic AI inside the IT stack, "which may not meet your security or quality standards." "Agentic AI will make decisions based on its analysis of your organization's data, making plans based on that analysis. From there, it'll act on those plans. This will be dangerous unless you invest in the skills, practices and technologies to deliver trustworthy AI agents. Your organization's data may be of poor quality, further increasing the risk. As well as creating risk, poor data quality and architecture will also inhibit agentic AI's development." Gartner also warned that AI agents could alienate customers if the experience is poorly designed. Gartner says organizations should create "customer journey maps to design the ideal customer experience and define guardrails before handing over to AI agents for execution." Whether it's a good idea or not, agentic AI is coming, Gartner says. It forecasts that a third of enterprise software will incorporate such agents while 20 percent of digital storefront interactions will be conducted by AI agents. "Agentic AI will be incorporated into AI assistants and built into software, SaaS platforms, Internet-of-Things devices and robotics," Gartner said. "When AI assistants start planning, making decisions, and taking action for you, agentic AI will be there. It'll be everywhere, with the potential to extend collaborative work management platforms beyond task tracking into planning and executing tasks." The benefit for organizations might be that - if employed properly - AI agents can "increase the number of tasks and workflows that can be automated." "The potential that agentic AI has to constantly analyze the performance of personalized interactions surpasses human capabilities, ensuring more precise and effective customer engagement. Software developers are likely to be some of the first affected, as existing AI coding assistants gain maturity and AI agents provide the next set of incremental benefits," said the research paper, Top Strategic Technology Trends for 2025: Agentic AI. Of course, risks can be managed for a reasonable cost if the potential benefits outweigh them. At this stage, though, we must ask ourselves whether any government or corporate IT department has ever been known to skimp on the hard part. ®

AI agents are coming -- game changer or just hype?

Hardly a day goes by at the moment without one company or another releasing a new AI agent system. Microsoft with Magentic-One and agents in Microsoft 365, Google Jarvis, OpenAI with its upcoming Operator and even the cautious Anthropic, have released or announced agent products over the past few weeks. It looks as though the world is about to face a deluge of AI agent systems over the coming two years, all vying to take over your everyday digital life. If your idea of an agent involves things like dry Martinis and fast cars, you're in for a surprise. Agents in the AI sense, are software routines that use artificial intelligence to scurry around and accomplish specific goals. They do this by splitting these goals into smaller tasks, using whatever resources and other AI agents they need, in order to monitor progress and complete the objective. Think of it like a group of very intelligent friends who get together to plan a fishing trip, led by their bossy friend -- the Ferris Bueller -- who coordinates them all, and makes sure that nobody goes off track at any time. Now you may ask, as some people do, why use agents when a standard AI model can easily do a lot of this drudge work itself? And that's a very good question. The fact is, while AI models are incredibly efficient at doing single tasks, they are not designed to do multiple things in a chain. Sure they can do complex work, if they're trained properly and given the right prompt. But they have their limits. Over 200 million users a week deploy ChatGPT for one task or another. But there's a huge difference between using and...well using. Looking at the numbers, it seems that most Jo Citizens are using AI as simple shortcuts in their work, almost like a glorified search engine, while the rest are enjoying a nice chat. Apart from spammers and scammers very few people seem to be using AI as a fully-fledged replacement for anything serious. Students use it to create essay outlines, and office drones summarize data, or 'delve' into banal snippets of marketing fluff. But nobody is sitting back, pressing an AI button and saying 'off you go'. There's always a human hovering around very close by, to make sure it's all working like it should be, and let's not talk about the horrors of hallucinations. Despite all the hype, the AI sector is still waiting for its killer application, the product that transforms a technology into a household tool, like spreadsheets and word-processing software did for business users around the globe decades ago. For the majority of us asking for a cheesecake recipe, it's hardly a revolution. Yet. Of course when AI shines, it really does bring a smile to the face. Specialist applications for professional users like programmers, medical researchers and financial analysts are showing just how powerful it can be to have a computer crunching data at unimaginable speed. So all this marketing noise around the newly arrived Agentic systems may be masking something a little bit deeper. The fact is, the pace of basic foundational AI development seems to have slowed up considerably. New model releases with incremental improvements are being trickled out, rather than the firehose we're used to. What is it that agents add that is so extra special? The reality is, in their present incarnation, they don't actually add that much to the mix. Sure they can make pretty charts, research the web and order a pizza, all while doing some clever math. But those are things that any AI model can do, and agents don't necessarily do them any better. In fact if you look at a current sample list of agents, and what they're capable of right now, you're likely to see the same old, 'summarize a piece of text or transcribe a YouTube video or conduct an SEO audit on a website', kind of stuff, which AI models have been doing for months, if not years. Basically software automation on steroids. Where things get interesting is when we start to look forward a little bit, to a time when there are other technologies in place which give agents the powers they need to differentiate themselves from 'dumb' AI models. One such example is Pin AI, not to be confused with Humane's troubled AI Pin. Pin AI, co-founded by ex-google DeepBrain alumni Bill Sun, is an agentic system which goes beyond simple goals and tasks, and implements a full ecosystem capable of accomplishing complex and sophisticated routines. At the heart of any truly revolutionary agentic system of the future, there will have to be some way of ensuring secure, private and trustworthy transactions out in the real world, without having to rely on human interaction. This is something that current AI models or agentic systems simply cannot do. Pin AI intends to address that problem by integrating AI with the blockchain, third party payment systems and composable smart contracts, which together will be able to accomplish a good percentage of the real-world transactions we humans need to do every day. If it comes off, it has the potential to truly revolutionize the way we do everyday tasks. So, instead of sending an AI agent off to just book us a cheap vacation trip on the web, we'll be able to ask an AI app to do the research on everything from cab fares to the airport, flights, hotels and holiday insurance. It will then use our trusted blockchain tokens to negotiate, book and manage our whole itinerary for the trip. Including booking and paying for a nice set of vacation activities based on our past history and known likes. Oh and hire the best tour guide on the island for the duration. This, and so much more, is the true future of empowered AI agents. Agents with full autonomy and the ability to connect with everything the real world has to offer. But for now, we'll have to wait patiently while the early adopters and enthusiasts work out the bugs and wrinkles of these fledgling systems, getting it all ready for prime time.

The AI race is shifting to AI agents. What are they?

A future where everyone has an artificial intelligence-powered assistant might not be too far off. While tech giants and AI startups have been racing to launch the smartest AI-powered chatbots, recent AI product rollouts point to a shift toward developing the most helpful AI-powered assistants -- or as some in the tech industry are calling them, AI agents. While addressing concerns over the impact AI will have on jobs, Nvidia (NVDA-1.32%) chief executive Jensen Huang said that it's likely every job, including his, will be transformed by the technology, because everyone will eventually have an AI assistant. Even Nvidia's chip designers are using AI assistants, Huang said during an interview at the industry conference SIGGRAPH 2024. Without AI assistants, the company's highly sought-after Hopper chips, which power some of the world's leading AI models, wouldn't have been possible, he said. The same apparently goes for the chipmaker's new AI platform, Blackwell. "I think the main reason is that the value that they capture and generate is just a magnitude different from what we've had so far," Daniel Vassilev, co-founder and chief executive of Relevance AI, told Quartz. "What agents are kind of promising us is this future where they can complete work that was previously impossible to be done with automation." Companies are starting to focus on AI agents because "everyone realizes now that we're in a place where the future is going to look very different -- we're going to have the capacity to do a whole lot more," Vassilev said. "I think a lot of people are realizing that the old way of doing things is probably also going to change." Simply put, an AI agent is software that can complete complex tasks autonomously. So far, most AI-powered tools have been co-pilots that can help users work more efficiently, Vassilev said. OpenAI's ChatGPT, for example, is a co-pilot. Users can interact with and work alongside ChatGPT, "but it's still ultimately an assistant" that has to be prompted over and over, Vassilev said. With AI agents, users can delegate work to the tool, then check to see if it needs assistance or if it has finished. Vassilev's company, Relevance, develops virtual workforces of AI agents that are used by major companies including Activision and Roku (ROKU+7.54%). Rebecca Greene, co-founder and chief technology officer of Regal, said AI agents are "like humans" that can be given personas and jobs to do. But, Greene said, AI agents are not meant to trick customers into thinking they're speaking to a real human. "When you're calling them or when they're calling you, they know exactly what they're trying to accomplish, what their goals are, what their personality is," Greene said. Regal, which develops purpose-built AI agents for call centers, believes 90% of interactions in contact centers are going to be AI-led over the next ten years, Greene said. For now, most AI agents are simple, Greene said, such as those used for setting up doctor's appointments. "An AI agent can place a phone call, not take the doctor's time, speak to the customer on the other side, find a mutually agreeable time to schedule something, book that appointment in the doctor's calendar, and the customer confirms it," she said. So far, Greene said Regal's clients see acceptance from customers for AI agents, which can perform better than some humans in areas such as speaking in several languages, being available at all hours of the day, and keeping composure in difficult situations. At SIGGRAPH 2024 in July, Meta (META+0.05%) chief executive Mark Zuckerberg announced the wider rollout of Meta's platform, AI Studio, which allows users and creators to generate AI characters of themselves that can act as an assistant for different tasks, including interacting with their community. Meta's vision, Zuckerberg said, is to empower users, from creators to small businesses, to create agents for themselves. In October, Microsoft (MSFT+0.20%) announced that its Copilot customers would soon be able to build their own "autonomous agents" in Copilot Studio, that can "understand" their work and "act" on their behalf. "These models offer more precise predictions, enhanced natural language processing, and improved decision-making support," Microsoft said. Meanwhile AI startup OpenAI is reportedly preparing to launch an AI agent, codenamed "Operator," that can do tasks on behalf of a person, including coding and booking travel, Bloomberg reported. The startup is also reportedly close to completing another AI agent project that can perform tasks in a web browser. As companies begin adopting AI agents, they will likely invest in building their own, Greene said, that align with their industry-use cases and that are trained on their data. Vassilev said the focus will be on less-risky parts of a business, such as highly-repetitive and monotonous tasks that workers don't want to do. Now that companies realize that "2025 is going to be the year of AI agents," Vassilev said he expects an uptick in AI agent products. Alongside that, people will have to build trust with AI agents the way they would with a human co-worker. "When you're hiring someone, you want to trust them, you want to know they're doing a good job," Vassilev said. "I think in a similar way, when you're adopting an agent that's going to be autonomous, you also need to build that trust."

Explainer: What Are AI Agents?

Rina Diane Caballar is a Contributing Editor covering tech and its intersections with science, society, and the environment. The artificial intelligence world is abuzz with talk of AI agents. Microsoft recently released a set of autonomous agents that could help streamline customer service, sales, and supply chain tasks. Similarly, OpenAI unveiled Swarm, an experimental framework to explore better coordination between multi-agent systems. Meanwhile, Claude, the large language model (LLM) from Anthropic, is taking agentic AI to the next level with the beta stage of its computer use skills -- from moving a mouse cursor around the screen to clicking buttons and typing text using a virtual keyboard. So, what exactly are AI agents? "AI agents are advanced artificial intelligence systems that are able to complete a task or make a decision," says Adnan Ijaz, director of product management for Amazon Q Developer, an AI-powered software development assistant from Amazon Web Services (AWS). "Humans set the goal, and agents figure out on their own, autonomously, the best course of action." The agents can interface with external systems to take action in the world. In addition to this autonomy, agentic AI can also receive feedback and continually improve on a task, says Yoon Kim, an assistant professor at MIT's Computer Science and Artificial Intelligence Laboratory. Think of AI agents as a more capable version of generative AI. While both technologies rely on LLMs as their underlying model, generative AI creates new content based on the patterns it learned from its training data. Agentic systems, on the other hand, are not only able to generate content but are also able to take action based on the information they gain from their environment. "So all of that is essentially a step further than generative AI," Ijaz says. To fulfill a particular task, AI agents usually follow a three-part workflow. First, they determine the goal through a user-specified prompt. Next, they figure out how to approach that objective by breaking it down into smaller, simpler subtasks and collecting the needed data. Finally, they execute tasks using what's contained in their knowledge base plus the data they've amassed, making use of any functions they can call or tools they have at their disposal. Let's take booking flights as an example, and imagine a prompt to "book the cheapest flight from A to B on Y date." An AI agent might first search the web for all flights from A to B on Y date, scan the search results, and select the lowest-priced flight. The agent then calls a function that connects to the application programming interface (API) of the airline's flight booking platform. The agent makes a booking for the chosen flight, entering the user's details based on the information stored in its knowledge base. "The key point of agentic interaction is that the system is able to understand the goal you're trying to accomplish and then operate on it autonomously," says Ijaz. However, humans are still in the loop, guiding the process and intervening when required. For instance, the flight-booking AI agent might be instructed to notify the user if the cheapest flight has no available seats, allowing the user to decide on the next step. "If at any point humans don't think the system is going in the right direction, they can override it -- they have control," Ijaz adds. Much like generative AI, agentic AI holds the promise of increased efficiency and improved productivity, with the agent performing mundane tasks that would otherwise be tedious or repetitive for the average human. "If these systems become trustworthy enough, then we could have agents arrange a calendar for you or reserve restaurants on your behalf -- do stuff that you would otherwise have an assistant do," says Kim. The keyword there is trustworthy, with data privacy and security as major challenges for agentic AI. "Agents are looking at a large swath of data. They are reasoning over it, they're collecting that data. It's important that the right privacy and security guardrails are implemented," Ijaz says. For instance, researchers at the University of California San Diego and Nanyang Technological University in Singapore were able to trick AI agents into improper tool use. They created a malicious prompt attack that analyzes a user's chat session, pulls out personally identifiable information, and formats it into a command that leaks the data to an attacker's server. The attack worked on Mistral AI's Le Chat conversational assistant, so the researchers disclosed the security vulnerability to the company, which resulted in a product fix. Factual accuracy is another issue for AI agents, since they're built on LLMs that have a problem with hallucinations -- the technical term for making things up. Kim notes that while it certainly wouldn't be desirable to have an AI agent give you the wrong information about flights, such a mistake probably wouldn't be disastrous. "That's not as high stakes an application as employing these types of systems in clinical or financial settings," Kim says, "where the accuracy or lack thereof of the outputs or actions could have serious consequences." Agentic AI is still in its early stages, and as AI agents evolve, they'll hopefully make people's lives easier and more productive. But caution is still recommended for the risks they pose. "It's an important advancement, so I think all the attention it's getting is warranted," Ijaz says. "Agents are another tool in the armory for humans, and humans will put those tools to good use granted that we build those agents in ways that follow responsible AI practices."

2025: The year 'invisible' AI agents will integrate into enterprise hierarchies

Join our daily and weekly newsletters for the latest updates and exclusive content on industry-leading AI coverage. Learn More In the enterprise of the future, human workers are expected to work closely alongside sophisticated teams of AI agents. According to McKinsey, generative AI and other technologies have the potential to automate 60 to 70% of employees' work. And, already, an estimated one-third of American workers are using AI in the workplace -- oftentimes unbeknownst to their employers. However, experts predict that 2025 will be the year that these so-called "invisible" AI agents begin to come out of the shadows and take more of an active role in enterprise operations. "Agents will likely fit into enterprise workflows much like specialized members of any given team," said Naveen Rao, VP of AI at Databricks and founder and former CEO of MosaicAI. AI agents go beyond question-answer chatbots to assistants that use foundation models to execute more complex tasks previously not considered possible. These natural language-powered agents can handle multiple tasks, and, when empowered to do so by humans, act on them. "Agents are goal-based and make independent decisions based on context," explained Ed Challis, head of AI strategy at business automation platform UiPath. "Agents will have varying degrees of autonomy." Ultimately, AI agents will be able to perceive (process and interpret data), plan, act (with or without a human in the loop), reflect, learn from feedback and improve over time, said Raj Shukla, CTO of AI SaaS company SymphonyAI. "At a high level, AI agents are expected to fulfill the long-awaited dream of automation in enterprises that robotic process automation (RPA) was supposed to solve," he said. As large language models (LLMs) are their "planning and reasoning brain," they will eventually begin to mimic human-like behavior. "The wow factor of a good AI agent is similar to sitting in a self-driving car and seeing it steer through crowded roads." However, AI agents are still in their formative stages, with use cases still being fleshed out and explored. "It's going to be a broad spectrum of capabilities," Forrester senior analyst Rowan Curran told VentureBeat. The most basic level is what he called "RAG plus," or a retrieval augmented generation system that does some action after initial retrieval. For instance, detecting a potential maintenance issue in an industrial setting, outlining a maintenance procedure and generating a draft work order request. And then sending that to the end (human) user who makes the final call. "We're already seeing a lot of that these days," said Curran. "It essentially amounts to an anomaly detection algorithm." In more complex scenarios, agents could retrieve info and take action across multiple systems. For instance, a user might prompt: "I'm a wealth advisor, I need to update all of my high net worth individuals with an issue that occurred -- can you help develop personalized emails that give insights on the impact on their specific portfolio?" The AI agent would then access various databases, run analytics, generate customized emails and push them out via an API call to an email marketing system. Going further beyond that will be sophisticated, multi-agent ecosystems, said Curran. For example, on a factory floor, a predictive algorithm may trigger a maintenance request that goes to an agent that identifies different options, weighing cost and availability, all while going back and forth with a third-party agent. It could then place an order as it interacts with different independent systems, machine learning (ML) models, API integrations and enterprise middleware. "That's the next generation on the horizon," said Curran. For now, though, agents aren't likely to be fully autonomous or mostly autonomous, he pointed out. Most use cases will involve human in the loop, whether for training, safety or regulatory reasons. "Autonomous agents are going to be very rare, at least in the short term." Challis agreed, emphasizing that "one of the most important things to recognize about any AI implementation is that AI on its own is not enough. We see that all business processes are going to be best solved by a combination of traditional automation, AI agents and humans working in concert to best support a business function." One example use case for AI agents that nearly every industry can relate to is the process of onboarding new employees, Challis noted. This typically involves many people, including HR, payroll, IT and others. AI agents could streamline and speed up the process as it receives and handles contracts, collects documents and sets up payroll, IT and security approval. In another scenario, imagine a sales rep using AI. That agent can collaborate with procurement and supply chain agents to work up pricing and delivery terms for a proposal, explained Andreas Welsch, founder and chief AI strategist at consulting company Intelligence Briefing. The procurement agent will then gather information about available finished goods and raw materials, while the supply chain agent will calculate manufacturing and shipping times and report back to the procurement agent, he noted. Or, a customer service rep can ask an agent to gather relevant information about a given customer. The agent takes into account the inquiry, history and recent purchases, potentially from different systems and documents. They then create a response and present it to a team member who can review and further edit the draft before sending it along to the customer. "Agents carry out steps in a workflow based on a goal that the user has provided," said Welsch. "The agent breaks this goal into subgoals and tasks and then tries to complete them." While agent frameworks are relatively new, some companies have been using what Rao called compound AI systems. For instance, business data and analytics company FactSet runs a finance platform that allows analysts to query large amounts of financial data to make timely investments and financial decisions. The company created a compound AI system that allows a user to write requests in natural language. Originally, the company had one monolithic LLM and "packed as much context as it could" into each call with RAG. However, this method hit a quality ceiling with around 59% accuracy and a 16-second latency, Rao explained. To address this, FactSet changed its architecture, breaking its system down into a more efficient AI agent that called various smaller models and functions, each customized or fine-tuned to accomplish a specific, narrow task. After some iterations, the company was able to significantly improve quality (85% accuracy) while decreasing costs and latency by 62% (down to 10 seconds), Rao reported. Ultimately, he noted, "true transformation will come from leveraging a company's data to build a unique capability or business process that gives that business an advantage over its competitors."

Autonomous agentic AI can shake up workflows, and businesses should prepare now

A new Deloitte report shows that agentic AI may be closer than you think. If you have used an AI chatbot, you have experienced firsthand how convenient it is to have an assistant who can answer any questions instantaneously. Now imagine if that assistant could understand and act on your needs without you telling it what to do -- meet autonomous agents or agentic AI. Although this vision may seem like something out of a sci-fi movie, many companies are working on making agentic AI a reality, with enterprise solutions already being released. Deloitte's 2025 TMT Predictions report predicts that 25% of companies that use generative AI will launch agentic AI pilots by 2025 and 50% by 2027. Also: Is Perplexity Pro worth the subscription? This free shipping perk just might convince me Furthermore, the report suggests investors have contributed over $2bn to agentic AI startups over the last two years, with efforts concentrated in the enterprise market. So, with such rapid growth and robust investments, how much will agents impact workers? Agentic AI refers to assistants that make decisions independently from human intervention, choosing what actions to take to accomplish a particular goal established by a human. They differ from copilots, which respond to human requests to act. Because agentic AI performs independently, the technology must be able to perform tasks reliably all the time -- and the technology isn't quite at the required level yet. Also: Agentic AI is the top strategic technology trend for 2025 The report says an agentic coding engineer, which would still require some human supervision, will be achievable by 2025. As seen in the chart below, a completely agentic coding engineer is so far away that there is no predicted date: The report also delineates how the impacts of agentic AI could be "enormous" because there are one billion knowledge workers globally and stagnant productivity growth in the US. Productivity had only increased by 0.5% from 2019 to 2023 compared to 0.8% growth from 1987 to 2023. Also: Gartner's 2025 tech trends show how your business needs to adapt - and fast Generic AI must be fully capable to help boost productivity, which the report suggests might not be too far off. "Most importantly, gen AI agents of all kinds need to be reliable for enterprises to use them: Getting the job right most of the time isn't enough. There are some use cases and applications in late 2024 that show encouraging signs of being reliable enough for adoption in early 2025," said the report. Agentic AI represents a new frontier for AI. However, the technology is built on pre-existing foundations, including large language models, enterprise applications, the internet, and multimodal capabilities. Also: I tested the cheapest Surface Pro 11 model: 3 main takeaways as a Windows expert Ultimately, the report suggests agentic AI will be valuable despite being at an early stage of development and adoption. The analysis says business leaders should prepare for the rise of AI agents by redesigning workflows and focusing on data governance and cybersecurity while maintaining skepticism. The report says areas of work that will benefit from agentic AI include customer support, cybersecurity, where there is a significant shortage of skilled knowledge workers, regulatory compliance, and agent builders and orchestrators.

AI agents' momentum won't stop in 2025

Join our daily and weekly newsletters for the latest updates and exclusive content on industry-leading AI coverage. Learn More One of the buzzwords of 2024 in AI has been agents, specifically the agentic future. AI agents have become one of the most talked-about trends for enterprises, and as more organizations look to implement agents, the future for agents may look rosy. In the next year, more enterprises could bring AI agents out of sandboxes and into production, making AI agents a big trend for 2025. Steve Lucas, CEO of platform integration company Boomi, said the conversation around AI agents picked up speed this year due to multiple factors in the growth trajectory of generative AI and models. "I believe there are moments in time in the course of history where there's convergence, and things come together to create an outcome we didn't expect so soon," Lucas said in an interview with VentureBeat. "You have near infinite compute, extraordinarily powerful GPU processing capabilities, data that is not near infinite, sprinkle in a fundamentally new way to take and process inputs and outputs that have all converged at the same time." In other words, AI agents became a big deal because we can see a path for these agents to actually work. When organizations and AI companies talk about an agentic future, they usually mean a time when many tasks within an enterprise are automated. People will either prompt, or do a simple action, and AI agents will begin fulfilling those requests. In the past few months, large service providers have begun offering access to agents to customers. Salesforce has gone all in on agents with the release of agents called Agentforce. Salesforce chairman and CEO Marc Benioff said during the launch of Agentforce that AI agents represent the "third wave of AI" which the company is very excited about helping to usher in. It isn't just Salesforce that is talking up AI agents. Slack will let customers integrate agents from Salesforce, Asana, Workday, Cohere, Writer and Adobe. ServiceNow updated its Now Assist platform with a library of ready-to-use agents, and AWS introduced Agents for Bedrock so clients can build custom agents more quickly. Lucas and other experts VentureBeat spoke to agreed that 2024 is the year enterprises realize they can bring agents into their technology stack. The following year will bring more agent deployment, but multiple agents working together could still take some time to work well. The momentum is not slowing down The various platforms available to access a library of agents or low-code ways to build custom agents make it easier for enterprises to consider using agents. The adoption of agents is already growing. A survey from Forum Ventures showed that among 100 senior IT leaders, it spoke to, 48% are ready to bring AI agents into operations. Around 33% said they are very prepared. As they continue experimenting and figuring out good use cases for their organizations, 2025 will allow companies to test out production in small tasks for agents. Deloitte Head of AI Jim Rowan said clients who've started limited tests of agents see the potential of agents "as skilled collaborators that enterprises have been searching for that understands personal preferences." Boomi's Lucas said his company is anticipating the number of customers using its agents "should go up 10x next year." He said around 2,000 clients actively use Boomi's agents. However, while 2025 could see a boom in agents, some enterprises may also consider the cost of using agents widely. Paul van der Boor, vice president for AI at investment company Prosus, told VentureBeat that agentic use will only keep growing, but companies have to remember there is a cost inherent to this technology. "The trajectory is not going change because I think the direction is clear," van der Boor said. "Keep in mind that there's also a lot of practical considerations because one of the things agents do is they require multiple calls to various elements, and they require more tokens, so they're more expensive." AI agents will see evolution, too Lucas said the best use of agents is when they move from solitary actors to digital employees working with each other and human workers to complete tasks. But we won't see multi-agents in production early in 2025. Lucas said what is most likely is the rise of agent islands. "You'll have islands, like the Salesforce island, the Boomi island, the Oracle island. Over time, these agents will talk to each other," he said. The next few years could see the rise of agents taking more of a proactive role in the enterprise. Deloitte's Rowan said some AI agents could become multipurpose agents that anticipate users' needs. For example, the agent could proactively scan someone's inbox, categorize inbound emails from clients, reference those with a list of priorities, tailor responses and flag any information to the employee. "Over time, agents will level up on the cognitive nature of the task they're performing. I don't think we're there yet because agents now are still operating more at the behest of the employee," he said. One future AI agent evolution could be a conductor or orchestration agent. Meta agents, one of the many terms for this concept, is an AI agent that directs traffic or actions of other agents. Paul Tether, CEO of market intelligence firm Amplyfi, said the so-called Meta Agents is the ultimate next step for enterprise AI agents. "By the end of next year, we'll start to see meta agents emerge," he predicted.