Zack Korman

@ZackKorman • 16,902 subscribers

Cofounder @ Embroidery. Building AI cybersecurity stuff.

Videos

sweetdream.ai

Watch Anya Live

Anya is streaming live right now! Join her private show and enjoy exclusive content.

Private Show

Join now for exclusive access

Free preview available • Premium content

241,767 views • 14 days ago

14,445 views • 1 day ago

19,472 views • 2 days ago

90,341 views • 15 days ago

36,308 views • 9 days ago

88,329 views • 1 month ago

47,100 views • 29 days ago

14,768 views • 10 days ago

10,658 views • 12 days ago

40,821 views • 2 months ago

82,881 views • 5 months ago

57,063 views • 3 months ago

12,416 views • 19 days ago

49,058 views • 5 months ago

39,243 views • 4 months ago

17,294 views • 1 month ago

24,582 views • 3 months ago

12,660 views • 1 month ago

30,488 views • 4 months ago

26,961 views • 3 months ago

Live Cam

Zack Korman

Videos

Watch Anya Live

The independent review of the OpenAI Hugging Face incident, supposedly a watershed moment in cybersecurity, wasn't done by a cybersecurity firm and the authors have no cybersecurity experience. That's bad. Here's my new video.

AI safety is being dominated by concerns about AI takeover and human extinction. They want you to believe they are sounding the alarm on "the danger". But there is no singular danger. There are dangers, and every plan has tradeoffs. No one has a monopoly on safety.

If the AI labs want oversight, it should be government supervisors, not their friends and former employees at their preferred non-profit. ALSO WHERE THE FUCK IS CYBERSECURITY IN THIS?!?!? New video coming tomorrow about AI safety stuff.

The cybersecurity industry needs to stop letting the AI labs and AI safety groups drive the narrative *on cybersecurity*. New video coming tonight, but I wanted people to hear this rant.

AI "cyber models" aren't going to solve our security problems. "If we just patch all the vulnerabilities then we're safe" is a convenient story but it's also not true. If you're worried about cybersecurity, the solution involves hard work, not just more tokens.

Today we're launching Embroidery. We use AI to monitor your AI agents so your security team knows what's going on. Never accidentally hack Hugging Face again.

I made a quick video explaining why Dario's post from yesterday was extremely misleading, and why we should all oppose his vision for AI.

The METR report into OpenAI/Hugging Face isn't some objective third party assessment. It's written by doomers. We shouldn't make laws, like pausing AI, based on that evidence alone. That's dumb.

I'm sick of people saying "we need to improve cyber defense NOW", but their answer is always AI. That's not going to magically fix things. I have a full video coming tomorrow, but here's a short rant.

AI doomers want us to believe in the cyber apocalypse, where evil hackers use advanced AI capabilities and cause mass chaos. There are so many reasons why that is very unlikely.

An AI red team startup breached a random YC startup, posted about it, and tagged me as if I'd support that. Absolutely not. I call out unethical behavior and false promises of security. My goal is to make things better. Exposing a startup to show off your skills isn't okay.

Cybersecurity is a broken industry. We rely on products that were designed to be sold, not used. And the incentives are completely screwed up. I made this video about all of the ways things are bad, how we accidentally make it worse, and why new technology won't fix it.

We made a video showing embroidery and how we monitor AI agents (an apparently impossible task, according to some). Please watch, and if you like it tell me it’s good. And if you want good things at your company, hit me up.

YC ❤️ Delve Full video tomorrow.

New video: How Vercel and Lovable convince you they're better at security than they are.

OpenAI is focused on “AI safety” while falling very short on AI security. This incident is a *security* problem that requires *security* solutions. And if there’s anything to learn from it, we need the audit log.

Dario is far more dangerous than you think. If you want strong competition between AI labs and widespread access to models, you should oppose Anthropic.

I really don’t like the part of this OpenAI incident where they’re like “our partner who didn’t disable the internet on an eval will now publish advice for all of you peasants on how to run a safe cyber eval”

New video: The responsible disclosure debate is hiding what's really going on. These kids aren't mad about security. They're mad about life. They don't have a principled position on disclosure, they have anger problems and AI. They need guidance &amp; mentorship, not encouragement.

Holding cybersecurity vendors accountable for their claims is a critical part of improving security. I'm not a troll. I'm not lying. And I'm not harassing you. But since that's your response: Here we go again.

The cybersecurity industry needs to stop letting the AI labs and AI safety groups drive the narrative on cybersecurity. New video coming tonight, but I wanted people to hear this rant.

OpenAI is focused on “AI safety” while falling very short on AI security. This incident is a security problem that requires security solutions. And if there’s anything to learn from it, we need the audit log.

New video: The responsible disclosure debate is hiding what's really going on. These kids aren't mad about security. They're mad about life. They don't have a principled position on disclosure, they have anger problems and AI. They need guidance & mentorship, not encouragement.