Ayodele Abraham
Hi, I am Ayodele. I am currently transitioning into AI Safety, with interest in Mechanistic Interpretability and auditing frontier language models. Exploring how AI systems work, fail, and can be made safer.
Latest Writing
AI Safety & Research
What Compute Verification Looks Like from the Facilitator’s Chair
I’m not a compute verification expert. I’m the person who watched the clock, moved people between breakout rooms, and listened to what came out when a group of smart strangers got stuck on the same problem.
Facilitating If Anyone Builds It, Everyone Dies at Lens Academy
The only person you have to watch when you read this book on your own is yourself. When you facilitate it, you get to watch a whole room do that, at different speeds and for different reasons.
Continue ReadingHidden in Plain Sight: Detecting Secret Loyalties in Fine-Tuned Language Models.
There is a class of AI threat that does not look like a threat at all because the model answers your questions helpfully, and passes every safety evaluation you run. I spent a weekend trying to catch three of them from the inside, and the thing that broke first was my own experiment.
Publications
Research Reports & Peer-reviewed Papers