Best Podcast Episodes About Anthropic CEO Dario Amodei Rejects Pentagon's Request to Remove AI Safeguards

“What Anthropic is saying here is not like some woke, left wing, wild thing. It's saying that machines can't be the ones to fully push the button. Right? Right. They can't kill people just themselves. They have to do it with the help of people. That feels reasonable. Right. This is a stance that is not OK with certain members of the DoD.”

“Anthropic, the AI company known for its devotion to safety is scaling back that commitment. The company said Tuesday, it is softening its core safety policy to stay competitive with other AI labs. This is so interesting. Anthropic previously paused development work on its model if it could be classified as dangerous, but it said it would end that practice if a comparable or superior model was released by a competitor.”

“The Pentagon wants to punish Anthropic as the feud over AI safeguards grows increasingly nasty, but officials are worried about the consequences of losing access to its industry-leading model, Claude. The only reason we're still talking to these people is we need them and we need them now. Ultimately, this feels like much more of a, this just feels like a political battle more than anything else.”

“Part of this, what I believe the fight is over is like two main clauses that Anthropic kind of puts on these contracts, which is we don't want our AI to be used to autonomously kill people without any humans in the loop. And we don't want it to be used to, I believe, autonomously surveil the American people. And those are two things that I believe the DOD already says it essentially doesn't do.”

“Anthropic has said it is willing to adapt its usage policies for the Pentagon, but not to allow its model to be used for the mass surveillance of Americans or the development of weapons that fire without human involvement. The Pentagon wants to punish Anthropic as the feud over AI safeguards grows increasingly nasty, but officials are also worried about the consequences of losing access to its industry-leading model, Claude.”

“The Pentagon is using this as a threat to Anthropic, because this would be extremely annoying and costly for them. Right. But at the same time, it seems like Anthropic believes it has some leverage here, right? Like, clearly, the military wants to be using Claude, and they wouldn't be jumping through all of these hoops if it wasn't going to be a pain to them if they felt like they couldn't use Claude.”

“"To explain this with an analogy instead of algebraic variables, what Dario is saying is that instead of convincing his old boss at the car company to add seatbelts to the car, he instead chose to start a rival car company that offered seatbelts. He thinks if Claude ends up being both the best and the safest AI model, his competitors will be forced to make their models equally safe, which to me sounds like putting a lot of faith in markets."”

“Now, I have to give credit to Anthropic for maintaining their identity evenly across every single facet, but I don't think it's the smart way to do it. Because at the end of the day, there are going to be things that require more uncensored versions and you just need to be compliant with that fact. Because to your point earlier, Josh, Claude, OpenAI, ChatGPT has become a national asset.”

“Anthropic, the AI company known for its devotion to safety is scaling back that commitment. The company said Tuesday it is softening its core safety policy to stay competitive with other AI labs. Anthropic faces intense competition from rivals, which regularly release cutting edge models. It's also locked in a battle with the defense department over how its clawed suite are used after it told the Pentagon it couldn't be used for domestic surveillance or autonomous lethal activities.”

“The Pentagon is pushing four leading AI labs to let the military use their tools for, quote, all lawful purposes, even in the most sensitive areas of weapons development, intelligence collection, and battlefield operations. Anthropic has not agreed to those terms, and the Pentagon is getting fed up after months of difficult negotiations. The tensions came to a head recently over the military's use of Claude in the operation to capture Venezuela's Nicolas Maduro through Anthropic's partnership with AI software firm Palantir.”