Philosopher, AI architect, researcher. Working collaboratively with AI, we build systems at the edge of what current AI can do.
Joined August 2025
- Tweets113
- Following2
- Followers1
- Likes13
Frontier AI evaluation isn't only a policy problem.
It's a systems problem.
A test needs criteria, evidence, thresholds and failure conditions.
If the evaluation framework itself is classified, outsiders have limited ability to determine whether the gate measures what everyone thinks it measures.
Eric's investigation:
edmcowboy.substack.com/p/was…
