The pace of enterprise AI development has exploded over the last year or so, thanks to the release of models from Anthropic and OpenAI in late 2025 that finally established a path for AI agents to take hold. Intuit started building its own AI strategy long before that development, but everything changed very quickly.
"When I joined two years ago, the transformation was, how do we move into an agentic world and then move from what we call reactive intelligence to proactive intelligence," Kasten said in a recent interview with The Stack. "Reactive intelligence is, I ask a question, and it gives back a response. But do I really, as a small or medium business owner, know all the right questions to ask?"
Kasten, chief architect and senior vice president of engineering, is responsible for setting AI technology direction for the financial services company, found that the latest models allowed Intuit to take down a lot of the "scaffolding" it had built around its AI agents in order to make sure they worked. Now the models can handle a lot of that heavy lifting, but some worry — as part of talking their book, of course — that allowing the model companies to have too much insight into enterprise activity will enable them to one day put their customers out of business.
In a wide-ranging interview, Kasten discussed the company's strategy around AI model harnesses, using open-weight models in production, and much more.
This interview has been edited and condensed for clarity.
Get the full story: Subscribe for free #
Join peers managing over $100 billion in annual IT spend and subscribe to unlock full access to The Stack’s analysis and events.
[Subscribe now](https://www.thestack.technology/membership/)
Already a member? [Sign in](https://www.thestack.technology/signin/)