Show Notes
The Equity crew was live at TechCrunch Disrupt 2026! Kirsten Korosec, Anthony Ha, and Sean O’Kane took over the Builders Stage Tuesday morning to tackle a growing challenge for the AI industry: what happens when AI agents go off-script, and who’s responsible for keeping them in check?
Recent incidents, from Anthropic pulling internal evaluations off the live internet after agents misbehaved to Microsoft CEO Satya Nadella calling for an “emergency brake” on AI models, have brought agent security into focus. The Equity team unpacked whether the problem calls for better engineering, new regulation, or both, and why startups are raising millions to secure AI agents.
Listen to the full episode to hear more about:
The Anthropic agent mishaps, including a false police report that went nowhere after landing in a spam folder
Nadella’s call for an AI emergency brake, and why the hosts wonder what took so long
The funding race to secure AI agents, from AIR’s $50 million raise to Reco’s $55 million and Oak’s $60 million for agent identity, plus Sequoia-backed Cymphony
Why independent verification and certification could be key to building trust, and whether Big Tech will buy its way out of the problem
Subscribe to Equity on YouTube, Apple Podcasts, Overcast, Spotify, and all the casts. You can also follow Equity on X and Threads.
Chapters:
00:00 Intro and what we’re excited for at Disrupt
01:34 Durability, security, and Anthropic’s rogue agents
04:41 Nadella’s “emergency brake” post
06:33 NVIDIA’s OpenShell and the engineering-vs-regulation debate
10:04 The agent security startup boom: AIR, Reco, Cymphony
13:38 When agents go rogue in real life
15:55 Certifications and standards
19:33 HiddenLayer and the jump to production
20:40 Will Big Tech just acquire these startups?
22:52 Consumer trust and agent adoption
25:05 Predictions: acquisitions vs. independent verifiers
27:26 What’s coming up at Disrupt
29:02 Outro Learn more about your ad choices. Visit megaphone.fm/adchoices
More from Equity
It seems like every big AI player is building a personal agent. Meta has Muse, OpenAI launched Dots, and a startup called Instinct recently raised $1 billion at a $10 billion valuation. But Eugenia Kuyda thinks the next generation of AI assistants needs to do more than book your restaurant reservations and find you hidden money in your Gmail.
Amulya Vishwanath, who spent seven years leading product marketing for NVIDIA’s robotics platforms, says robotics is currently in its “Lego era,” with the basic building blocks like actuators, robotic hands, software, and safety standards still being rebuilt.
Amazon says it will stop using NDAs when negotiating data center deals with local governments, following a similar move from Microsoft earlier this year. Secrecy has fueled community backlash against AI infrastructure, with opposition leading to hundreds of proposed and enacted moratoriums from New York to San Francisco. Meanwhile, a wave of startups is betting that consumers will hand AI agents access to their inboxes, files, and credit cards — if they can get websites to let them in.
Tristan Walker and Brynn Putnam both built and sold major consumer companies. Now, they’re building products designed to bring people together.
This week, the White House got nearly every major tech CEO in one room — Zuckerberg, Bezos, Musk, and Anthropic’s Dario Amodei among them — to sign an AI safety pledge that President Donald Trump called “morally binding.” Trump also signed an executive order officially rebranding AI as “super intelligence,” and meanwhile, Meta and OpenAI are putting friendlier faces on their AI products, even as the biggest money in AI still appears to be coming from the enterprise.