Dario Amodei, CEO von Anthropic, weist darauf hin, dass OpenAI „die Risiken, die sie eingehen, nicht wirklich versteht“
Anthropic CEO Dario Amodei suggests OpenAI doesn’t „really understand the risks they’re taking“
Dario Amodei, CEO von Anthropic, weist darauf hin, dass OpenAI „die Risiken, die sie eingehen, nicht wirklich versteht“
Anthropic CEO Dario Amodei suggests OpenAI doesn’t „really understand the risks they’re taking“
27 Kommentare
From article
>“We’re buying a lot. We’re buying a hell of a lot. We’re buying an amount that’s comparable to what the biggest players in the game are buying,“ Amodei says. „But if you’re asking me, ‚Why haven’t we signed $10 trillion of compute starting in mid-2027?‘ First of all, it can’t be produced. There isn’t that much in the world. But second, what if the country of geniuses comes, but it comes in mid-2028 instead of mid-2027? You go bankrupt.“
Amodei says he gets the impression that some competitors „don’t really understand the risks they’re taking. They’re just doing stuff because it sounds cool,“ adding that Anthropic has „thought carefully about it.“ While he only refers to „some of the other companies,“ the comment reads as a likely jab at OpenAI.
Don’t understand or don’t care
They’re all psychotic, including Dario, let’s not pretend otherwise
Do not unleash a pure optimizer into the wild.
Not only ClosedAi, all of them anthropic included don’t fathom what they are doing, shooting in the dark at its finest
If their actions dont have consequences then there are no risks. As long as everyone keeps throwing money at them then they wont stop asking for it, why would they stop?
Pot calling Kettle, black. Lol.
Of course they know and he knows they know.
Rich asshole slap fight.
A competitor criticizes a competitor?! What is the world coming to??
Ohhh they do understand it. They just don’t give a fuck
We may have to be taxing companies by token usage. You use more AI, you pay more into the treasury. I think it’s the only way to not end up in an America where you have to step over 10 homeless families to get to work.
That said, we are going to fail this test and it’s going to be a nightmare. We do NOT have the government in place which would amortize the damage of the new Industrial Revolution.
They are not going to fix Capitalism – and so they will get Socialism, and in this case, it’s the only way.
God these people are so full of themselves. They made an LLM that can help programmers code and maybe trick old people with fake news. They act like the Terminator is just around the corner to hype their chatbot.
„Oooohh it’s so good it’s *scary*“
Meanwhile the energy use is enormously impactful to the climate and Anthropic is the primary vender for the DoD, but Anthropic doesn’t talk about that risk.
and anthropic is taking the „do no evil“ marketing mask. we’ve seen it before.
Every single one of these AI companies is attempting to intertwine themselves so far into American government that when it all collapses (it’ll 100% collapse) then they’ll get a trillion dollar bailout and then they’ll exist literally forever propped up by the damn feds. Banks did this. Real estate did this. Now AI will do it.
I don’t think Anthropic understood the risks they were taking when they made the brilliant decision to partner with Palantir. Fucking hypocrites.
All these companies are a disease to the existence of all living things.
Something, something hyperbolic. News echoes. Refresh in people’s minds as they doomscroll. Rinse and repeat.
Welcome to modern tech news.
They really do though. So factor that into the „evil“ equation.
They do. They just don’t care because to them they’re not risks, but opportunities. For profit, for control.
Nothing like one AI company telling another AI company they don’t understand AI risk.
consider the latest chinese AI thats clearly trained on popular movies, the output isnt any better than a typical Marvel movie. Is it time to ask if the current approach to AI has a ceiling of what its trained on?
More and more people are using AI these days, with some even starting to spend more time with it than in meaningful human interactions. As long as this trend continues, the thirst for deeper-thinking and faster-responding AI won’t stop.
Regarding this conclusion, I disagree with Dario. I’m not saying he’s incorrect, but I believe the balance between supply and demand will remain skewed. Otherwise, we face a future where only those who can afford high monthly fees have the privilege of using AI.
None of these companies care. In an effort to be first they are ignoring all safety standards and precautions
funny how safety concerns get louder right when your competitor is about to close a bigger deal. every single time.
This whole drama reminds me of Binance FTX relationship
Kettle calls pot black, news at 11