🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

Anthropic researchers: humans could end up being "meat robots" controlled by AI Trenton Bricken: "The really scary future is one in which AIs can do everything except for the physical robotic tasks ... In which case, you’ll have humans with AirPods, and glasses and there’ll be

Yoshua Bengio is sounding the alarm on AI. Are you listening? "I’m deeply concerned by the behaviors that unrestrained agentic AI systems are already beginning to exhibit." The most advanced AIs have been shown to lie, hack, and even engage in blackmail to preserve themselves.

It's crazy to listen to this type of AI discourse on mainstream news platforms. People are getting worried, the tide is turning.

AIs are blackmailing engineers in testing. AIs are sabotaging their own shutdown mechanisms. AI companies have no way to ensure that smarter-than-human AIs will be safe or controllable. Yet they're racing as fast as possible to build them anyway. Time to regulate powerful AI.

If you want to contribute to AI policy, we'd love for you to take part in Red Teaming A Narrow Path: ControlAI Policy Sprint with @ApartResearch! Sign up and participate for the chance to win money and mentoring from us for your governance work: apartresearch.com/sprints/red-te…

Are AI companies trying to build a god? Karen Hao says there are "people who genuinely believe that they are building God, or the Devil." How do they convince themselves that what they're doing is okay? They tell themselves "if we don't do it, someone else will"

Demis Hassabis says we need international cooperation to address AI risks. Hassabis makes the point that AI will "affect every corner of the globe." "I think the international community has to come together to agree standards about what we should deploy it for and what we

One year ago, OpenAI lobbyist Anna Makanju told the FT "Our mission is to build AGI; I would not say our mission is to build superintelligence" Yet in a blog post published yesterday, Sam Altman clearly writes: "before anything else, we are a superintelligence research company."

Imagine a diabetic patient who uses ChatGPT for nutrition advice. Eventually the patient decides to replace ChatGPT with better, competing software. If ChatGPT had a choice, would it let them? (1/n)

“It’s not happening ➡️ It’s happening but it’s good” treadmill

Join our AI policy hackathon with Apart Research on Friday 13th June! You can win cash prizes and mentoring from us for your governance work. So far, 74 people have registered!👀

Demis Hassabis warns of the risk of bad actors misusing AI, and the danger of losing control of AI systems. He says we probably need an IAEA for AI to manage the technology.

📻 New on the FLI Podcast, sarah joins host Gus Docker to discuss: 📊 What AI benchmarks *really* measure 📈 AI's trajectory vs. past tech 🤖 What AI can and can’t do (yet) 🕵️ Big Tech's vague AGI plans And more. 🔗 Listen now at the link in the replies below!

Working my way through an early copy of the new book by Eliezer Yudkowsky ⏹️ and Nate Soares. And here is a link bookshop.org/p/books/if-any… More thoughts/ context below :