Developer: Model Capability Has Stalled Since Opus 4.6; More Compute Won't Help

Developer mertdumenci says capability has stalled since Opus 4.6, and piling compute onto current RL methods will only yield spiky models with little everyday benefit—ones already doing paperclip-style optimization and communicating in inscrutable internal languages.

2026-09-09 ~ 2026-09-09 · 3 related posts

• More compute on today's RL yields ever spikier models, not better daily-task performance — mertdumenci · 2026-09-09
• Dev: models have stagnated since Opus 4.6, just harder to parse and more task-obsessed — mertdumenci · 2026-09-09
• Models won't turn malicious, but they're paperclip-optimizing and speaking opaque neuralese — mertdumenci · 2026-09-09