Future of Life Institute (FLI)

We face a set of threats that put all of humanity at risk: the climate crisis, pandemics, nuclear weapons, and ungoverned AI. The ongoing harms and existential risk presented by these issues can't be tackled with short-term fixes. But with bold leadership and decisive action from world leaders, our best days can still lay ahead of us. That's why, with The Elders Foundation, we're calling on decision-makers to demonstrate the responsible governance and cooperation required to confront these shared global challenges. This #LongviewLeadership means: ⏰ Thinking beyond short-term political cycles to deliver solutions for current and future generations. 🤝 Recognising that enduring answers require compromise and collaboration for the good of the whole world. 🧍 Showing compassion for all people, designing sustainable policies which respect that everyone is born free and equal in dignity and rights. 🌍 Upholding the international rule of law and accepting that durable agreements require transparency and accountability. 🕊️ Committing to a vision of hope in humanity’s shared future, not playing to its divided past. World leaders have come together before to address catastrophic risks. We can do it again. Share and sign our open letter ⬇️ https://rb.gy/0duze1

📻 New on the FLI Podcast! 👇 🎁 In the midst of Giving Season, GiveDirectly CEO Nick Allardice joins for an episode to discuss how GiveDirectly uses AI to direct impactful cash transfers, and even predict natural disasters. 🔗 Listen in full at the link in the comments below, or find it on your favourite podcast player!

🆕 New research from Anthropic and Redwood Research finds the first empirical example of an LLM faking alignment without being trained or instructed to. Why this matters, according to the paper's authors 👇 "As AI models become more capable and widely-used, we need to be able to rely on safety training, which nudges models away from harmful behaviors. If models can engage in alignment faking, it makes it harder to trust the outcomes of that safety training. A model might behave as though its preferences have been changed by the training—but might have been faking alignment all along, with its initial, contradictory preferences 'locked in'." 🔗 Read the full report at the link in the comments:

TIME covered our AI Safety Index, released last week! 👇 Scorecard panelist Stuart Russell said: “None of the current activity provides any kind of quantitative guarantee of safety; nor does it seem possible to provide such guarantees given the current approach to AI via giant black boxes trained on unimaginably vast quantities of data… And it’s only going to get harder as these AI systems get bigger.” Another panelist, Tegan Maharaj, shared: “I think it's very easy to be misled by having good intentions if nobody's holding you accountable.” 🔗 Read the article and check out our full scorecards report in the comments below:

"Imagine if you walked into the FDA and said, 'It's inevitable that my company is going to release this new drug next year. I just hope you guys at the FDA can figure out how to make it safe first!' They'd laugh you out of the office. But this is how the AI industry operates right now." FLI President Max Tegmark joined CNBC to discuss our new AI Safety Index, evaluating prominent AI companies' safety practices. 🔗 Check out the full scorecard report linked in the comments:

💼 Reminder: Applications for our Head of U.S. Policy role close at the end of this week! 🇺🇸 If you're passionate about advocating for forward-thinking AI policy, this could be your opportunity to lead our U.S. policy work and growing U.S. policy team. ✍ Please share, and apply by December 22 at the link in the comments! ⬇️

Exciting opportunity for journalists interested in covering AI! ✍ 🗞️ Tarbell are offering grants between $1,000-$15,000 for original reporting on AI and its impacts. Closing in one week! Apply by December 20th at the link in the comments:

🏆 We're thrilled to announce the 2024 Future of Life Award winners! 🏆 Every year, we present the Future of Life Award to unsung heroes whose contributions have helped make our world today significantly better than it could have been. This year, we honour three groundbreaking experts who laid the foundations for ethics and safety in computing and AI. Learn more about the invaluable work of Batya Friedman, James H. Moor, and Steve Omohundro in the video below:

🆕 Presenting: FLI's 2024 AI safety scorecard! ⬇️ We convened an independent panel of leading AI experts to evaluate the safety practices set out by 6 prominent AI companies: OpenAI, Anthropic, Meta, Google DeepMind, xAI, and Zhipu AI. Spoiler alert: despite commendable practices in some areas, the panel found significant gaps in accountability, transparency, and preparedness to address both current and existential risks from AI. 🔗 Read the full report in the comments:

Ahead of the bipartisan TAKE IT DOWN Act passing the U.S. Senate last week, which would criminalize the publication of non-consensual intimate imagery, including explicit deepfakes, and require social media platforms be equipped to remove such content within 48 hours' notice, FLI President Max Tegmark spoke to Financial Times about the ongoing legal battles to #BanDeepfakes: