Shortly after Ashok Srivastava ascended to the role of chief AI officer at Intuit over the summer, he combined the artificial intelligence unit with the “futures” team, which is tasked with understanding and accelerating adoption of AI and other emerging technologies like quantum computing and blockchain.
This newly formed team is made up of a few hundred employees across several geographies, including New York, California, and Texas, as well as international offices in Israel and India. The group is made up of the VPs for the futures and AI/ML (machine learning) divisions, software engineers, AI scientists, product managers, and AI and ML engineers.
Called Intuit Foresight, the team’s mandate is to stitch together AI research and practical applications of AI that can be deployed to support the work of the financial technology company’s developers, who can then launch new AI functionality for Intuit’s customers.
“We fundamentally believe that in order to deliver the very best customer outcomes, we need to have the most advanced capabilities,” says Srivastava.
One example of his thinking that predates the combined Intuit Foresight team—but points to Srivastava’s rationale—is his team’s critical role in one of Intuit’s new AI agents for accounting. The agent debuted this summer as part of a collection of AI agents specialized in tasks like financial analysis and payments, which Intuit CEO Sasan Goodarzi said was the “most significant launch we’ve ever had in our history.”
Using open-source AI, Srivastava’s team created foundational large language models that he says are highly accurate but with low latency, which is the time delay between when an input is received and when the AI system is able to generate an output.
Srivastava says that this accounting agent is able to save customers an average of 12 hours per month and achieves an accuracy rate of above 90% for the transactions it categorizes. Intuit says transaction categorization is a repetitive task, but enables accountants or business customers to monitor and make adjustments as needed. Over time, the system applies learnings from that human feedback for future transaction categorization tasks.
“When you take that dozen hours saved for a small business or a mid-market business, it’s a very significant saving when you add it up and multiply it across all the businesses that we’re supporting, which are in the millions,” says Srivastava.
So far, Intuit’s big bet on AI has paid off. In November, the company reported better-than-expected fiscal first-quarter results owing to growth from its AI products, which have proliferated across a portfolio of brands that includes TurboTax, QuickBooks, and Credit Karma. Last month, Intuit also struck a $100 million multiyear contract with OpenAI, bringing some of its platform capabilities directly into ChatGPT.
Moving forward, AI and generative AI will continue to be the top priority for the Intuit Foresight team. And Srivastava points to a lot of progress that the company has already made on generative AI. Since the AI agents launched in July, more than 2 million customers within Intuit’s business platform have engaged with these tools, with repeat engagement of over 80%. Intuit’s technologies have created 3,500 different generative AI use cases internally and externally throughout fiscal 2025. Within the IT department, generative AI tools have resulted in 40% faster average coding, leading to 39% more code delivered per developer.
Srivastava says that even with all that progress, there remain more opportunities to improve.
“We need models that can do the very best reasoning,” he says. “We need models that can do the very best forecasting and scenario-planning usage of tools that are specialized to the financial domain. We are seeing the benefits that I mentioned earlier because of these techniques.”
When asked if other large organizations should carve out a dedicated “lab” to focus on AI development, Srivastava offers some words of caution. “If companies don’t have their data in order and don’t have an AI platform that they built or they bought, and they don’t have the talent to take advantage of what’s possible today, it’s premature to create a team like this,” says Srivastava. “We’ve been working on this for nearly a decade.”
With hundreds of employees working in Intuit Foresight, the increasingly expensive war for AI talent is also top of mind for Srivastava. Intuit says the salaries it offers are “highly competitive,” based on continuous benchmarking against market rates and updated when necessary.
But Srivastava stresses that the technologists he courts aren’t only motivated by finances, even though Intuit is itself a financial tech company. He says there’s a broader mission to serve individual customers and small and midsize businesses that’s attractive to prospective employees. He says he’s shocked by how many applicants he sees for roles on his team, both within Intuit and externally, calling the demand “unprecedented.”
“We’re very competitive, but my sales pitch is really, ‘Come and spend your time and your energy making the world a better place,’” he says.
