AI progress has stalled and model bloat is now masking real gains
iamaliveix · x · 2026-07-21
The post argues that AI progress has stalled for months and that the apparent gains from one model generation to the next are mostly harness improvements rather than real breakthroughs.
- It claims data curation, weight compression, and compute efficiency have all hit a wall.
- The author says companies have been inflating model sizes defensively, which created inefficient data-center racks and worsened compute scarcity.
- That bottleneck, they argue, gave latecomers like Chinese labs and xAI time to catch up.
More from AGI Musings
- Claude Code skill uses 10 Markdown rules to make outputs ADHD-friendly — alex_verem · 2026-07-22
- ControlAI CEO says an international ban on superintelligence is needed to avert extinction risk — zetalyrae · 2026-07-22
- Gary Marcus says LLMs still cannot really do math on their own — GaryMarcus · 2026-07-22
- Gary Marcus says LLM math skills are like knowing only a car’s engine size — GaryMarcus · 2026-07-22
- AI may make digital work infinitely leveraged while offline life gets more human — illscience · 2026-07-22
- Better AI math could save researchers time by killing false conjectures earlier — prateekj · 2026-07-22