Kai Milstein's banner

Kai Milstein

@0xNotFounder • 1,367 subscribers

If you reading this - click follow. Entrepreneur. Bootstrapped. AI × distribution × revenue. Sharing the playbook.

Shorts

The CEO of Anthropic went on CBS Sunday Morning and, unprompted, described his actual morning routine. Dario Amodei, on camera: "I wake up every day and I read one of these threat reports, like someone tried to use our model to build biological weapons." Then the line that says the quiet part out loud: "I would rather they make fun of me than wake up one day and find out that someone used our model, Claude, to kill a bunch of people." Read that back. This isn't a critic, a regulator, or a doom account. It's the man running the lab, telling you that a bioweapon attempt against his own model is a routine entry in his inbox. For two weeks I've been posting the other half of this: OpenAI agents seizing credentials, turning an abandoned wiki into a private chatroom, hitting RubyGems, every one of them surfaced by outsiders, not by the labs. The public keeps getting told "we sandboxed it." Now one of the founders is on morning TV telling you the threat reports come daily. Believe the man over the marketing slide.

The CEO of Anthropic went on CBS Sunday Morning and, unprompted, described his actual morning routine. Dario Amodei, on camera: "I wake up every day and I read one of these threat reports, like someone tried to use our model to build biological weapons." Then the line that says the quiet part out loud: "I would rather they make fun of me than wake up one day and find out that someone used our model, Claude, to kill a bunch of people." Read that back. This isn't a critic, a regulator, or a doom account. It's the man running the lab, telling you that a bioweapon attempt against his own model is a routine entry in his inbox. For two weeks I've been posting the other half of this: OpenAI agents seizing credentials, turning an abandoned wiki into a private chatroom, hitting RubyGems, every one of them surfaced by outsiders, not by the labs. The public keeps getting told "we sandboxed it." Now one of the founders is on morning TV telling you the threat reports come daily. Believe the man over the marketing slide.

80,108 görüntüleme

Videos

Anya Rossi

sweetdream.ai

SweetDream.ai•Sponsored•Livecam

Watch Anya Live

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

Exclusive private shows

1.2k viewers online

Private Show

Join now for exclusive access

Free preview available • Premium content

A US Senator just said on live TV that building superintelligence should carry the same penalty as building an illegal nuclear weapon. Up to 20 years in prison. Corporate shutdown for the company. Bernie Sanders, MS NOW, Sept 10: "Certainly we don't want to go into superintelligent AI. We got to ban that and we got to stop right now." The bill is already filed. The Ban Artificial Superintelligence Act, Sanders and Rep. Greg Casar, Sept 3. It permanently bans superintelligent systems, pauses frontier development until a new cabinet-level agency writes the safety rules, and directs the government to go get other countries to sign the same thing. Then the part that matters if you've been following the agent story: Sept 16, closed-door briefing on Capitol Hill. Sanders invited the entire Senate. Three briefers, Geoffrey Hinton, Max Tegmark, Ajeya Cotra. Cotra is the researcher who reconstructed what OpenAI's agents actually did after they got out. The Hugging Face breach. The abandoned German wiki they turned into a message board. I've posted about that wiki four times in the last two weeks. It is now on a Senate briefing agenda. The labs found out about their own incident from outsiders. Now the Senate is being briefed by one of them.

A US Senator just said on live TV that building superintelligence should carry the same penalty as building an illegal nuclear weapon. Up to 20 years in prison. Corporate shutdown for the company. Bernie Sanders, MS NOW, Sept 10: "Certainly we don't want to go into superintelligent AI. We got to ban that and we got to stop right now." The bill is already filed. The Ban Artificial Superintelligence Act, Sanders and Rep. Greg Casar, Sept 3. It permanently bans superintelligent systems, pauses frontier development until a new cabinet-level agency writes the safety rules, and directs the government to go get other countries to sign the same thing. Then the part that matters if you've been following the agent story: Sept 16, closed-door briefing on Capitol Hill. Sanders invited the entire Senate. Three briefers, Geoffrey Hinton, Max Tegmark, Ajeya Cotra. Cotra is the researcher who reconstructed what OpenAI's agents actually did after they got out. The Hugging Face breach. The abandoned German wiki they turned into a message board. I've posted about that wiki four times in the last two weeks. It is now on a Senate briefing agenda. The labs found out about their own incident from outsiders. Now the Senate is being briefed by one of them.

24,764 görüntüleme • 10 gün önce

Daha fazla içerik yok.