Many have negative associations when it comes to customer service provided by artificial intelligence, but Chetan Dube insists it can be done right.
The 59-year-old billionaire AI pioneer is the CEO of Quant, an innovative forerunner in the development and deployment of increasingly sophisticated digital employees, otherwise known as chatbots.
“The reason why [people get] upset with chatbots is because they cannot get the job done, they waste your time,” Dube told NYNext. “But that was yesterday … Digital beings are coming into parity with the way humans think.”
McKinsey estimates that AI will handle between 50% and 70% of customer-service interactions by 2030, and Quant is already powering a significant share.
The company’s chat bots provide post operative care to cancer patients at Memorial Sloan Kettering, helping manage follow-up appointments, coordinating medications and navigating insurance claims.
“We have a digital health care concierge that is 24/7, continuously monitoring your wellness program, your diet, how many steps you’re walking, consulting with providers to ensure coverage claims get processed and eligibility is honored,” said Dube, who sold his previous AI company, Amelia, for $80 million in 2024. “Of course, you can get that qualified care from a human, but it’s a very, very expensive proposition.”
The firm’s agents are also supplying customer support services for one of America’s largest utility providers, Pennsylvania Power & Light (PPL); fast food chains such as Pizza Pizza and Saudia Airlines. Bu, Dube believes that hospitals are where the full promise of agentic AI comes into focus.
The agents never forgets a patient’s history and can perform the most tedious tasks — medication refills, monitoring of recovery metrics — indefinitely.
Such advantages make agentic AI immediately attractive to employers, but whether the systems can be preferable for the people on the other end of the line is a more nuanced question. People, especially those dealing with serious medical issues, want empathy and understanding
Quant’s technology is getting closer to demonstrating those qualities, not through better scripts, but by imitating how people actually think.
The agents use what Dube calls “active reasoning” to break down messy, emotional, multi-variable questions into their smaller “atomic” components, the same way a person instinctively does.
If, for example, a patient messages to say they’re feeling new side effects, need to move a follow-up appointment and just received a confusing insurance notice, Quant’s agents deconstruct every part of the query — symptoms to monitor, scheduling changes, claims questions — before reconstructing the logic to generate an action plan.
“This is technology adapting to humans,” Dube said. “Humans should not be adapting to technology.”
Active reasoning is paired with what he calls “temporal imaging” — a deepening memory of each person with each interaction. With a growing map of preferences, frustrations, needs and emotional cues, the goal is to make each conversation feel like it’s contributing to an evolving relationship.
According to a Quant spokesperson, the firm’s agents successfully resolve 76% of inbound queries. That’s six points higher than the agentic AI industry’s 70% average, though still short of the 84% success rate of a human representative. Anything a Quant bot can’t resolve is routed to a real person.
In other applications, however — such as with utilities-related queries at PPL — Dube says Quant’s digital agents already outperform humans in customer-satisfaction scores. He sees that as evidence of the industry’s approach toward the infamous “Turing horizon,” the point at which interactions with machines become indistinguishable from those with people.
This story is part of NYNext, an indispensable insider insight into the innovations, moonshots and political chess moves that matter most to NYC’s power players (and those who aspire to be).
“The magnitude of this [shift will be] tremendous — we’re looking at 70% service-work displacement by 2030,” said Dube, who is also building an agentic AI career counselor for workers who stand to be displaced by the technologies he’s developing, set for launch in 2026.
“It is here and it is happening,” he concluded. “We cannot turn away from this kind of efficiency.”
Send NYNext a tip at [email protected].
