FriskAI raises $3.6M from MaC Venture Capital to watch what AI agents do
Tech Funding News Abhinaya Prabhu ● Covered by 4 sources
FriskAI just raised $3.6M to log what AI agents do after they’re deployed. The bet: companies won’t trust agents until they can see every move they make.
Based on reporting by Tech Funding News, Abhinaya Prabhu — read the original for the full story.
Summary, retelling and take written by AI under human oversight; images are AI-generated illustrations. How we work · Report an error
FriskAI has come out of stealth with $3.6 million in pre-seed funding to track what AI agents actually do once they’re live. MaC Venture Capital led the round, with backing from Wischoff Ventures, NEA partner Rick Yang and a group of strategic angel investors. The company is aiming at healthcare, insurance and finance — the places where “we think it worked” is not a comforting answer.
The pitch is simple and, frankly, overdue. Enterprises have spent the past year giving agents access to databases, internal tools and customer systems, then discovering that very few monitoring products can explain what those agents did with that access. FriskAI sits between an agent and the tools it uses, logs each action, and gives security and compliance teams a record they can actually read.
Founder and chief executive Neel Palrecha says current monitoring still assumes software follows a script. That’s the old world. FriskAI’s idea is that agents need runtime intelligence — visibility while they’re working, not a postmortem after something goes sideways. Jason Moore, a staff software engineer at Sana Benefits, is already using the product and says it helps the team troubleshoot faster and deploy AI with more confidence.
The market around this is getting crowded quickly. Manifold raised $8 million for endpoint security for agents. Geordie AI raised a $30 million Series A for agent governance. NeuralTrust, Oasis Security and InfiniteWatch are all chasing adjacent problems, from identity to customer-facing observability. FriskAI is betting that a plain record of what an agent did is the thing buyers want first, before they care about fancier layers on top.
Palrecha founded the company in 2025 in Los Angeles after working on production systems at Apple, Headspace, Snapdocs and PlayVS. MaC Venture Capital’s Marlon Nichols framed the round as part of a bigger infrastructure cycle, the kind that creates the companies everyone else ends up depending on. The startup’s $3.6 million will go toward engineering and go-to-market work, which sounds right; trust in agents won’t be won by a demo alone.
My take — AI-written commentary, not fact-checked reporting
This is the sensible kind of AI company: boring in the best way, and built for the part of the stack people keep skipping. Europe has been shouting about AI safety and accountability while the money mostly chases shiny agent demos; FriskAI is proof the unglamorous tooling is where the real bill comes due. Closed-model hype gets the headlines, but audit trails and runtime visibility are what enterprises will actually pay for.
Read more about this at: Tech Funding News