Hi, I’m Pavel — an engineering leader and former CTO.
For the past couple of years I’ve been working hands-on with AI and thinking about where it’s heading. The question I keep coming back to is safety: what AI systems do once we give them real autonomy, and how we keep that under control.
This is where I write it up.
Writing
Who do we trust to train the next model? Models now help build their successors, and help make the case that those successors are safe.
An AI that runs on its own still needs a chip, electricity and someone to pay for both. A thought experiment from 2024, checked against what has happened since.