Axios

WASHINGTON, D.C. – Small businesses see big potential in AI, but getting the most return on investment depends on model quality, trust and safety, tech and business leaders said.

• Axios' Scott Rosenberg and Nathan Bomey moderated the roundtable discussion in Washington, D.C., on Sept. 17. The event was sponsored by Intuit.

Why it matters: AI can be a force multiplier for small- and midsize businesses (SMBs), allowing them to automate processes and operate more effectively.

5 big takeaways from the conversation:

1. SMBs jumping straight to customer-facing, rather than testing AI on internal operations to iron out the kinks first, could see a lower rate of return, UCLA chief data and AI officer Chris Mattmann said.
2. Automation is the payback most businesses are interested in, but it's still in the early stages, Veris AI CEO Mehdi Jamei said. "There's exhaustion from trying different tools that promise but don't deliver. … We need another six to 12 months from the technology side to be sure enough for people to see the actual advantage right now."
3. People have FOMO, so they want to use AI models that are ready to go, AI Squared VP of federal technology Napoleon Paxton said. "Large language models, ChatGPT shows … we can actually get into production-like things and not think about all the complexity."
4. Large enterprise software companies will need to build a trust layer between data and AI agents to gain buy-in from SMBs, Salesforce global public sector EVP Paul Tatum said. "What's interesting with these large language models is…they're non-deterministic. … The more guardrails you give them, the more confused and frustrated they get."
5. A national privacy standard "can go a long way" in establishing a foundation for data management practices, said Ashley Berrang, the Information Technology Industry Council's EVP of public affairs. It could also help small businesses that lack the resources to manage the patchwork of state laws across the country.

Sponsored content:

Intuit chief privacy officer Elise Houlik highlighted that during this time of AI innovation, small businesses need to ensure they have "access to good, quality tools that can help them."

Companies like Google should pay up front for their patent applications as AI supercharges the work of inventors, Google general counsel Halimah DeLaine Prado told Axios in an interview.

Why it matters: Google is sending a signal to John Squires, the newly confirmed leader of the U.S. Patent and Trademark Office.