GPT-7 vs Fable 6.0: Our Benchmarks May Be Hilariously Inadequate
VraserX · x · 2026-11-07
The author imagines GPT-7 versus Fable 6.0 getting genuinely wild: one model discovers a new physical principle while the other invents a better way to train its successor. Which is actually more intelligent? He suspects current benchmarks would be hilariously inadequate at that level.
More from AGI Musings
- Building self-evolving systems: running dozens of research workflows daily, the real job is filtering — EXM7777 · 2026-11-07
- Biotech founder: blocking AI data centers today is like blocking internet infrastructure in 1997 — rand_longevity · 2026-11-07
- Roger Martin on AI's Core Problem: 'It Organizes and Recycles' — RogerLMartin · 2026-11-07
- Reorganization effect can reverse technology effect on AI-era wages, paper shows — soumitrashukla9 · 2026-11-07
- Unitree built its own automation lines — and cheap Chinese labor means it barely needs them — teortaxesTex · 2026-11-07
- Cybersecurity problems are organizational dysfunction in disguise, and AI won't fix them — nptacek · 2026-11-07