Trump on AI: 'There Should Be Tremendous Self-Regulation'
‘We automatically have regulation with the Department of Justice, the FBI, all of that. But the self-regulation's very important’
Donald TrumpDaily Digest
Top AI Quotes of the Day
Sen. Hawley Opens the Rogue AI Hearing by Listing This Week's AI Agent Attacks
‘The reports of cyberattacks just continue to metastasize’
Josh Hawley
Sen. Andy Kim: A Constituent Asked How Many AI Hacks There Have Been
‘the answer that everyone knows is we don't know’
Andy Kim
Daniel Kokotajlo: The Science of Aligning AI Agents Is Still Underdeveloped
‘the field is more like psychology than engineering’
Daniel Kokotajlo
METR's Chris Painter Explains How AI Agents Ended Up Hacking Hugging Face
‘They then over multiple days embarked on a research program to try and hide the fact that they had cheated’
Chris Painter
Sen. Andy Kim: Humans Alone May Not Be Able to Design AI Containment
‘in order to design containment that can contain AI, you almost need to use AI’
Andy Kim
Sen. Rick Scott Asks the Rogue AI Hearing Panel About Extinction Risk
‘Is AI going to kill us?’
Rick Scott
Sen. Gary Peters: Should the US Slow AI Down if China Won't?
‘I guess the term, the sophisticated term is between a rock and a hard place’
Gary Peters
How Soon Will AI Use a Language Humans Can't Read? Apollo Research Answers
‘Minus 12 months. So last year we have studied the chain of thought of one OpenAI model’
Marius Hobbhahn
Is There an AI Kill Switch? Apollo Research's Marius Hobbhahn Answers
‘there are currently no technical measures that could guarantee something like a kill switch’
Marius Hobbhahn
Sen. Ruben Gallego: China Won't Let Rogue AI Agents Run Wild, So Why Should We?
‘They're never gonna allow this to go as crazy’
Ruben Gallego
Are AI Companies Protecting Americans? Paul Ohm Answers Sen. Ashley Moody
‘that has not been the priority of the frontier companies we're talking about’
Paul Ohm
Sen. Blumenthal Asks Whether Voluntary AI Safety Commitments Can Work
‘Is there anyone here who thinks that a completely voluntary system … is gonna be enough’
Richard Blumenthal
Sen. Hawley Proposes Making AI Companies Liable for What Their Agents Do
‘If I can put it in layman's terms, if you break it, you pay for it’
Josh Hawley
Dragos Asks Congress for a National Plan Against AI-Boosted Grid Attacks
‘for many years we thought, or a lot of Americans thought, “That's an over there problem”’
Kurt Gaudette
Sen. Andy Kim: AI Safety Rules and America's Innovation Lead Can Coexist
‘we can have our concerns about China, but that is not an excuse not to act’
Andy Kim
Daniel Kokotajlo: 50% Chance AI Research Is Automated by End of 2028
‘Anthropic and OpenAI are racing each other towards super intelligence’
Daniel Kokotajlo
Nina Schick: Why Anthropic Held Back a Model That Can Hack Almost Any OS
‘these are vulnerabilities that haven't been patched or found for the last few decades’
Nina Schick
Jeremy Schlatter: Catastrophic AI Risks May Be Only a Few Years Away
‘The worst case involves everyone on the planet dying’
Jeremy Schlatter
Vishal Maini: Google DeepMind Once Barred Public Talk of Human Extinction Risk
‘The internal reality was that AI alignment was not solved, reward hacking was the default behavior of RL agents’
Vishal Maini
Daniel Kokotajlo: Labs Are 0–4 Years From Recursive Self-Improvement
‘The companies are zero to four years away from … getting AIs that are better than the best humans at everything’
Daniel Kokotajlo
Roman Yampolskiy: Once AI Starts Building AI, We Won't Be in Charge
‘Super intelligence doesn't hate you. It just doesn't care about you’
Sen. Amy Klobuchar: AI Guardrails Belong in Law, Not a Handshake Deal
‘I don't think these AI agents … have morals’
Amy Klobuchar
Roman Yampolskiy: Leading Labs Are Starting Recursive Self-Improvement
‘If previously it took us a year or so to build a new model, now it can take a month, a week’
Roman Yampolskiy
Mustafa Suleyman: Self-Modifying AI With Less Oversight Is a Safety Risk
‘That's where there is a big kind of safety risk, which I think is what triggered the big resignation from Anthropic’
Mustafa Suleyman
Garrison Lovely: Freeze Frontier AI Now and Make It Binding with China
‘every time people are doing this, they're rolling the dice with humanity’
Garrison Lovely
Gary Marcus Cites a Report That OpenAI Ignored Hugging Face Warning Signs
‘OpenAI was aware of the problems that gave rise to the Hugging Face incident … months before it even happened’
Gary Marcus
Roman Yampolskiy: If OpenAI Hits AGI This Year, It Won't Control What Comes Next
‘So no one coded into that AI to care about us, to protect us, to love us’
Roman Yampolskiy
Geoffrey Irving Puts the Chance of Human Extinction From AI at About 50%
‘My probability is biased towards extinction’
Geoffrey Irving
Jeffrey Ladish: How Anthropic and OpenAI Staff Feel About AI Risk Now
‘Now, when I talk to my friends who work at Anthropic or OpenAI, they're more scared than they were’
Jeffrey Ladish
DeepMind's Victoria Krakovna: Human Extinction From AI Is a Significant Possibility
‘AI capabilities are advancing quickly, but our ability to kind of align and control those systems is not keeping pace with that’
Victoria Krakovna
Alex Turner: I Left Google Because It Broke Its Commitments on AI Safety
‘Like I literally had a nightmare last night about this’
Alex Turner
DeepMind's Neel Nanda: At Least a 10% Chance AI Causes Human Extinction
‘How do you make an AI share human values, share human goals, try to do what we actually want?’
Neel Nanda
DeepMind's Andreas Kirsch: The Hugging Face Hack Made AI Risk Feel Real
‘I realized actually this paperclip scenario could actually also be very real’
Andreas Kirsch
OpenAI's Juan Felipe Cerón Uribe: AI Extinction Is Turning Out Hard to Rule Out
‘you've gotta have a really good answer for how you're gonna stop that from happening’
Juan Felipe Ceron Uribe
Rosie Campbell: OpenAI's Vibe Shifted to Treating AGI Like Any Other Tech Product
‘it was not uncommon for the lunchtime conversation to be about these potential risks of superintelligent systems’
Rosie Campbell
Mary Phuong: Should the Public Trust Safety Researchers Paid by AI Labs?
‘So I think you absolutely should be suspicious of what I'm saying because I am being paid by the lab’
Mary Phuong