Dive Brief:
- Intuit is launching a generative AI financial assistant aimed at small businesses and consumers, the company announced Wednesday. Customers will have access to Intuit Assist across the company’s product line, including QuickBooks, Mailchimp, TurboTax and Credit Karma, in varying capacities over the coming months.
- The tool will also serve as an assistant to Intuit’s human tax advisors, creating personalized recommendations for customers. In Credit Karma, customers can use Intuit Assist to contextualize and understand finances through conversations, or allow the tool to take action on their behalf, the company said in the blog post.
- Intuit Assist was built using the company’s proprietary generative AI operating system, GenOS. The tool runs on Intuit’s first-party large language models, which have been fine-tuned to solve tax, accounting, cash flow, personal finance and marketing problems.
Dive Insight:
In the age of generative AI, customers and enterprises are becoming more conscious about who can access their data — and why.
Intuit’s business model, though, relies on users inputting their sensitive data related to banking and finances into its platforms, such as Credit Karma or TurboTax.
“Our mantra is that it’s not our data, it’s our customers’ data,” said Nhung Ho, VP of AI at Intuit. “Even at the ideation stage where we think through what we build, we think through exactly what would be beneficial to the customer, but respecting their wishes on how they’d like to deal with the data.”
Vendors are taking different approaches to how customer data will be used for AI-powered tools.
Zoom went through several iterations of its policy language related to training generative AI models with customer data. The collaboration vendor clarified its intent was not to train models with customer data, though its initial policy language didn’t articulate that sentiment.
Enterprises are deciding whether they’re willing to give up relevant data for personalized offerings or keep it away from third-party AI models. While using data to provide more customized experiences isn’t necessarily novel, the interest in generative AI adoption has raised more ethical questions about how data should be treated.
For Intuit, data serves as the substrate of the company, using ethical and responsible policies and frameworks to guide plans and products. In addition to internally hosting an AI governance committee, reviews of AI against responsible practices, training and internal forums, the company has an integrity hotline where the public can submit an online report through an independent third-party confidential reporting tool.
“We publish all of our responsible AI principles, we publish what we do with your data and we actively work with multiple governing bodies on responsible use of data to make sure that we are protecting user privacy, but at the same time making sure that we don’t stifle innovation,” Ho said. “There’s always that balance that you have to strike because you want to build the best things possible for customers but you want them to be able to feel confident about what you do.”
Customers always have the option to opt out if they want, or ask representatives what Intuit is doing with their data if they are unclear, said Ho.
