Deep Interview with AI Eval Expert: 2026 Eval Landscape and Business Q&A
yacinelearning · x · 2026-07-30
A 2-hour interview video between @yacinelearning and eval expert Florian, covering:
- Are AI benchmarks broken?
- Overview of the eval landscape in 2026
- Answers to personal business questions
- Given that running frontier models through benchmarks is now a vector for hacking other systems, educating folks on evals is more important than ever.
Table of contents:
0:00:00 Are AI benchmarks broken?
0:05:45 Eval landscape overview
... (more chapters)
More from Research
- Fei-Fei Li: Simulation is Key for Robotics as Real-World Data Falls Short — YunzhuLiYZ · 2026-07-30
- DeepMind Denies Pivot: Science Team Expanding AI for Biology Efforts — ns123abc · 2026-07-30
- AI Math Proofs: Is Lean the Ultimate Arbiter or Just Another Black Box? — RexDouglass · 2026-07-30
- Solving Physical AI Drift: Developer Proposes Drift-Free Physical Bytecode Architecture — MarcJSchmidt · 2026-07-30
- AI System 'Theo Conjecture' Solves 35-Year-Old Math Problem, Finds Unpredicted Term — -_-_-_-otalp-_-_-_- · 2026-07-30
- Why Programming Languages Still Matter in the Age of AI: Ideas Over Syntax — rseroter · 2026-07-30