Now, Zoom's bet on the future of work, Zoom AI Companion 3.0, spotlights a potential maturation of that existing paradigm. The company released Wednesday (Sept. 17) a set of agentic artificial intelligence capabilities, including AI-generated avatars that can autonomously represent employees in meetings.
But are agentic AI avatars coming to your next B2B sales meeting?
Digital sales assistants are not new. Chatbots, email sequencers and predictive lead scorers have long automated slices of the B2B sales process. But these were tools and extensions of the human salesperson.
What distinguishes agentic AI avatars is agency, or the ability to initiate actions, hold conversations and adapt strategies in real time without explicit, moment-by-moment instructions from a human overseer.
B2B sales and lead-gen teams are watching this development closely.
For enterprise sales teams, the economics may turn out to be compelling. A single account executive can only manage so many client conversations per week before cognitive fatigue and scheduling bottlenecks set in. An AI avatar, by contrast, scales elastically. In theory, a firm could "staff" hundreds of simultaneous discovery calls, nurturing prospects across time zones and languages at negligible marginal cost.
At the same time, B2B teams risk the threat of agentic avatars turning out to be merely artificial polish. And nothing is more deadly to a sales target than one-size-fits-all inauthenticity.
Read also: Agentic AI Puts a Face on Corporate Treasury's Next Leap
For decades, the B2B sales funnel has been a linear construct with awareness at the top, interest in the middle and purchase at the bottom. It was a simple but powerful model, useful for guiding sales teams through a predictable series of steps. Today, that funnel has become less of a neat diagram and more of a dynamic ecosystem.
Technology, ranging from data analytics to AI and automation platforms, is reshaping how companies attract, engage and convert customers. In the process, the idea of the sales funnel is being redefined. Agentic capabilities may just be the next step in this timeline.
The risk, however, lies in over-automation. Buyers are quick to detect inauthentic outreach, and AI-driven personalization must strike a careful balance between customization and human empathy. Technology can augment sales, but it cannot replace trust.
B2B is built atop trust and nuance. Can an AI stand-in persuade a chief financial officer to approve a six-figure payments solution, or reassure a procurement lead about compliance and risk? Firms might be hesitant to find out.
Still, more than 1 in 10 enterprise CFOs are now using or testing agentic AI technology to transform their planning, operations and decision making. The PYMNTS Intelligence report "How Agentic AI Went From Zero to CFO Test Runs in 90 Days" found that 52% of these firms are building in-house capabilities with internal AI and engineering teams.
See also: Executives Using AI Avatars Face a Patchwork of Regulations
Delivering on this future vision of agentic B2B commerce is non-trivial. AI avatars must master a delicate mix of natural language understanding, real-time rendering and context-sensitive reasoning. The stakes are higher than casual chatbots; sales conversations are nuanced, with subtle cues, objections and relationship dynamics. An avatar that fumbles a key industry acronym or mishandles a skeptical prospect could undermine a company's reputation.
There's also the potential for heightened fraud. AI has made fraud more challenging to battle, according to the PYMNTS Intelligence report "Rising Risk: Confronting Modern AP Fraud Threats." AI-generated deepfakes and impersonations have become the top threats for accounts payable teams.
The European Union's Artificial Intelligence Act and the Federal Trade Commission's guidelines on deceptive marketing could extend to business communications, particularly if avatars impersonate humans or obfuscate their machine nature.
Integration into existing customer relationship management and marketing systems is another prerequisite. For businesses to trust avatars, their outputs must flow smoothly into Salesforce dashboards, email cadences and performance analytics. Otherwise, the AI salesperson risks becoming a silo rather than a value multiplier.
The trajectory may mirror the adoption of chat-based sales assistants a decade ago. Initial skepticism will give way to experimentation, followed by selective integration in industries where speed, scale and global reach outweigh the premium placed on human connection. Software as a service, digital infrastructure and eCommerce platforms may lead adoption; professional services and high-touch consultative sectors may lag.
Ultimately, the question is not whether avatars can technically participate in sales calls, but whether they should. Business interactions are more than data exchanges; they are rituals of trust, empathy and negotiation. Even the most advanced AI may struggle to replicate the chemistry of a handshake, the intuition of a seasoned account manager, or the subtlety of reading a boardroom's mood.
