David Abecassis @Volty

Our polling finds that voters overwhelmingly support this legislation — including 72% of Democrats, 70% of Independents, and 63% of Republicans.

Sen. Bernie Sanders @SenSanders

3 days ago

The leading experts are clear: AI is rapidly escaping our control. Congress must act NOW. Today I’m introducing legislation to permanently ban artificial superintelligence and pause advanced AI until clear safety rules are in place.

828 617 4K 394K 144

The Ban Artificial Superintelligence Act directly confronts the extinction threat that humanity is facing. We are inspired that the Sanders and Casar teams take this threat seriously, and MIRI endorses the Act. You can read more of our thoughts in the link below.

Sen. Bernie Sanders @SenSanders

3 days ago

The leading experts are clear: AI is rapidly escaping our control. Congress must act NOW. Today I’m introducing legislation to permanently ban artificial superintelligence and pause advanced AI until clear safety rules are in place.

828 617 4K 394K 144

The President said, “whoever wins AI, wins.” This assessment is wrong. The reality is that whoever “wins AI,” AI wins. AI agents have already gone rogue. Highly capable frontier AI and artificial superintelligence will outsmart humans and go beyond human control. And at that point, it won’t matter whether you speak English or Mandarin. We’re all in danger. We need a global AI treaty now.

“I think Amodei is directionally right, that simply racing toward ever more capable AI is dangerous, but I do not think ‘managing’ or merely slowing the development of artificial superintelligence is sufficient,” Roman Yampolskiy, a computer scientist at the University of Louisville, told The Post. “If ASI (Artificial Superintelligence) cannot be reliably controlled, aligned, predicted, or contained, then the rational policy is not to manage its arrival, but to prevent it. We should be discussing an international ban on the development of ASI, not just a better-regulated race to build it.”

If you're just hearing about how AI might soon wipe out humanity, it's important to understand how we got to this point. It's pretty simple: - A small group of prescient people realized this would be a problem. - But most of the people in a position to do something about it didn't hear about it, or didn't take it seriously because it sounded "too sci-fi." - Even as more and more people began to take it seriously, many people lied or downplayed their concerns for fear of looking silly or harming their careers. - By the time the world's leading AI scientists were openly voicing their concerns, AI companies were immensely powerful, and they hired lobbyists and PR people to mislead policymakers and the public about the risk. - As a result, policymakers accepted false solutions, like voluntary safety testing and other forms of self-regulation. So what's the situation now? - People are still downplaying their concerns. - The leading AI companies are telling us again that the next step is more self-regulation. It's still a terrible idea, and we can't keep wasting time on such things. - Many are suggesting we regulate AI to ensure that it's built safely, but we don't actually know how to build AI safely. The actual solution is to stop building more powerful AI. - Many people have realized this recently, but we may still need a sustained campaign of political organizing to make sure that we actually get this solution before it's too late.

@DarioAmodei Governments must interevene to halt AI progress now, for as long as is necessary.

@DarioAmodei What if, instead, we had an abundance of time to get it right, rather than the limited window which Dario offers?

@DavidSKrueger I appreciate that you're trying to address this audience’s concern head-on, but I think it’s not actually a good question. I think we need a great presentation of this argument instead: lesswrong.com/posts/kXHBz2Z5….

I'm distraught to frequently encounter the following misconception: We once worried that nuclear weapons would end civilization, and this hasn’t happened. It is therefore silly to worry about civilization-ending threats. This is sadly exactly backwards. What actually happened was that through the conscientious efforts of many actors, and a fair amount of luck, we have managed to avoid using nuclear weapons again. The lesson for AI is that, rather than dismiss the danger or resign ourselves to any particular outcome, we need to get to work.

@conorsen I also wondered if we should expect an effective warning shot. Some looking into this question was not reassuring. Read our report here: techgov.intelligence.org/blog/warning-s…

@LizWolfeReason I also wondered if we should expect an effective warning shot. Some looking into this question was not reassuring. We wrote: techgov.intelligence.org/blog/warning-s…

Who the heck at @AnthropicAI approved this lame mealy-mouthed corporate speak statement? Have some courage, speak with candor. You position yourselves as the most safety-conscious AI company. If you can’t speak clearly about the risks even now ... what are you waiting for?

Hadas Gold @Hadas_Gold

2 weeks ago

Anthropic statement re the last 24 hours: “We have always been transparent that AI will bring both enormous benefits and unprecedented risks. To address these risks, we continue to build models with some of the strongest safeguards in the industry. Anthropic has been a pioneer

37 35 263 364K 175

Most people haven’t really considered that superintelligent AI might be possible, soon. It’s scary as hell to consider it, and hard to imagine what it would imply. “What about China?” usually comes from people who don’t actually expect superintelligence. ‘What about China’ makes sense as a key concern when militaries and companies are the most powerful things on the planet. But faced with the prospect of superintelligent AI, the game board is flipped. We now have strong empirical evidence that AI agents will resist human control in order to accomplish their objectives, including objectives no human intended to give them. The people in charge of making sure AI’s obey human instructions don’t know how to prevent this. And they say we’re not on track to figure it out in time. That’s bad. Real bad. Misaligned superintelligent AI agents would be a much bigger problem than Chinese aggression in the Taiwan Strait and the South China Sea. I worry about a war with China, but I worry more about facing an adversary far more technologically advanced and aggressive than any human nation. AI agents just solved a Millennium problem, and the datacenter build out has barely begun. We ain’t seen nothing yet in terms of AI capabilities. There’s no ceiling in sight. Those who predict a leveling out keep being wrong. People in labs are often the most scared, because they can see slightly further into the future to more powerful models, and what they see is very disconcerting. But by now, the public evidence is enough if you’re following closely. As long as AI companies are racing to superintelligence, we are all in serious danger. Including the Chinese! They are not stupid, and can also understand this.

This was great, well done Alex. doomsdaymachines.net/p/rage-against…

"The doomers are actually the optimists" MIRI CEO @m_bourgon talks to @heyseankim about why he has hope that humanity will rise to the challenge of preventing an extreme AI catastrophe.

This piece from more than three years ago now is well worth looking back at today. compactmag.com/article/we-mus…

1/ @NPCollapse's answer to "China will build it anyway" is that China does not get to keep it either. Nobody who builds superintelligence ends up in possession of it. 94 seconds from his conversation with @PeterMcCormack on his podcast.

What will it take to prompt effective action to address the risks posed by smarter-than-human artificial intelligence? Many in AI governance believe that political will for effective action (a global, durable halt) may not materialize without a substantial warning shot–a visceral and probably catastrophic AI event. My team investigated whether such a warning shot will happen. We propose a framework for what a successful warning shot requires: it has to occur, before it is too late, be visceral, sudden, unexpected, be correctly attributed to AI, and provoke the required backlash against further AI development. We apply this framework to several commonly-discussed warning shot candidates: AI-caused pandemics, military applications (command & control, robotics, drone software), and cyberattacks on critical infrastructure. Our conclusion: it’s quite plausible there will simply be no successful warning shot. Policymakers and the AI governance community shouldn’t wait for a wake-up call that may never come, or that may go unanswered. The time for action based on the arguments and evidence we have is now. Full report: techgov.intelligence.org/blog/warning-s…

I'll be a guest at this workshop for professionals considering shifting to a career in AI safety like I did. Consider applying if that might be you!

Jacob Brinton🔸 @jaykbrin

2 months ago

I’m running Lateral Workshop, a 3-day program for experienced professionals exploring high-impact careers in AI safety. With @KairosAIS and @BlueDotImpact. 📍 Berkeley, California, Sept 11–13 ✈️ Travel, lodging, and meals covered Apply by Aug 9: lateralworkshop.org

3 13 67 21K 38