Updated our 2022 paper on the alignment problem (ICLR). More direct evidence now supports our hypotheses: alignment faking, manipulative and obfuscated reward hacking, situational awareness, internal goals and power-seeking. Key updates below. 1/🧵
arxiv.org/abs/2209.00626 Just yesterday* OpenAI confirmed one of hypotheses: obfuscated reward hacking. GPT knowingly hacks its reward function ‘in the wild’; OpenAI penalizes it; so GPT keeps doing it in secret.


*not even in the paper yet

https://x.com/OpenAI/status/1899143752918409338

Tired of waiting 💤 while your model trains? Try skipping points that are already learned, not learnable or not worth learning! Robustly reduces required training steps 🏎 by >10x ! to reach the same accuracy on big web-scraped data

📜ICML 2022 paper: arxiv.org/abs/2206.07137 Training on big web-scraped data can take ages 💤 But lots of compute and time is wasted on redundant and noisy points that are already learned, not learnable, or not even worth learning.

“Understanding the effectiveness of government interventions in Europe’s second wave of COVID-19” -- a result of a new collaboration between @Oxford, @Imperial, @FHIOxford, @OATML_Oxford, @uni_copenhagen, @LSHTM, @Cambridge_Uni, @TheCrick... 1/

➡️Paper: medrxiv.org/content/10.110… As Europe enters a third wave of COVID, policy-makers balance controlling infections with the sweeping socioeconomic costs of interventions. To do so, we must know how effective individual interventions were at controlling COVID. 2/

Excited to share a major update on how effective 8 interventions have been against COVID-19 transmission in 41 countries.

Paper: medrxiv.org/content/10.110… Work done with great colleagues from 13 research groups, supervised by @yaringal, @yeewhye, Leonid Chindelevitch. Currently in submission.