Alignment Jam: Safety Benchmarks June 30 – July 2 2023

Join us to hack away on research into AI safety benchmarks! Please RSVP: https://forms.gle/arLyFrcJNvQJjn3H9

Prague is joining more than 30 locations around the world to research the safety of ML systems as a part of the Safety Benchmarks Hackathon.

Large AI models are released into the world by the month. We need to find ways to evaluate models (especially at the complexity of GPT-4) to ensure that they will not have critical failures after deployment, e.g. autonomous power-seeking, biases for unethical behaviors, and other inversely scaling phenomena.

Participate in the Alignment Jam on safety benchmarks to spend a weekend with AI safety researchers to formulate and demonstrate new ideas in measuring the safety of artificially intelligent systems. You will compete with participants from across the globe and get a great chance to review each others‘ projects as well!

We especially welcome students, researchers, and practitioners of Machine Learning and related fields.

Check out provided resources on the topic:
– https://openai.com/research/safety-gym
– https://aypan17.github.io/machiavelli/
– https://github.com/inverse-scaling/prize

Click here to display the event on Facebook