Acorns just acquired Frich, a social-finance app for Gen Z with 1.6 million users — and is shutting the app itself down.
Most acquisitions are bought to keep the product running. This one was bought for the people who used it, not the thing they used.
"This is a growth and customer-acquisition play, specifically by moving earlier... the hardest part of fintech is frequently customer acquisition." — Maia Bittner
That's Maia Bittner, Chief Product Officer at Imprint, on this week's Fintech Insider Podcast by 11:FS, alongside Confide founder and CEO Pav Gill and Huntington National Bank's Vijay Konduru, Head of Business Innovation & AI Products. The panel also covered a new AI governance platform built to watch both humans and AI agents, and a live argument over whether a chatbot now beats a human on the phone.
The full episode is covered here so you can skip it. 56 minutes of audio, 7 minutes of reading.
Here are the 4 takeaways that matter.
Fintech Insider Podcast · Oct 26, 2026 · 56m · 49m saved
With Pav Gill, Founder and CEO of Confide · Vijay Konduru, Head of Business Innovation & AI Products at Huntington National Bank · Maia Bittner, Chief Product Officer at Imprint
Key Takeaways
Confide wants one audit trail covering every employee and every AI agent, Pav Gill says — a user can ask it how many bribery cases ran in the last ten months and get an answer in seconds
Pav thinks automating that oversight could eventually spawn its own AI liability-insurance category, as accountability concentrates on whoever's left doing the work
Maia Bittner says a good AI chatbot is already the superior customer-service experience — Pav Gill disagrees, and says he'd rather press a thousand buttons to reach a human
Huntington built an entire AI operating system, 1866 Studio, just for small-business owners who have to be their own marketer, accountant and operator in the same day
Acorns bought Frich for its 1.6 million Gen Z users, not its app, which Acorns is shutting down, Maia says — Pav warns the same peer-comparison model can slide into status-flaunting or hidden debt
1. One Audit Trail For All
Pav Gill, who founded the whistleblowing-software-turned-compliance platform Confide, described its new agentic governance, risk and compliance product as an attempt to pull case management, policies, risk and audits into one system that tracks both human employees and AI agents.
Confide's pitch is a single audit trail for every action a human or an AI agent takes
"One of the biggest enablers of governance failures is data fragmentation... you still have very, very advanced big companies relying on Jira and Slack and Excel spreadsheets just to do investigations." — Pav Gill
Pav framed the problem as fragmentation: departments build around their own software and don't communicate across tools, so even large companies still run investigations by hand. Confide's answer, he said, is a plain-language question a user can ask directly.
"You could literally speak to Confide and ask it, how many bribery cases have you had in the last ten months, how did you deal with them — and it just turns it out in seconds." — Pav Gill
Vijay Konduru, who runs AI products at Huntington National Bank, said the bank's own approach rests on three principles: engineering for trust with guardrails and evaluation frameworks, matching the degree of autonomy to the use case, and keeping a human in the loop because the models themselves are non-deterministic. Maia Bittner agreed AI suits this kind of manual investigative work but flagged hallucination risk in a governance context, arguing a narrowly purpose-built tool mitigates it. Pav added that as routine compliance work gets automated, accountability concentrates on the people who remain — which he thinks could eventually produce AI-specific liability insurance, similar to professional indemnity cover.
