Follow-up testing: GPT 6.2 stacks jargon on long requests, 'lacking theory of mind' for its reader
johnowhitaker · x · 2026-11-20
johnowhitaker reports on overnight tests of GPT 6.2 with longer, more ambitious requests: results lean heavily on terminology that makes no sense without context, showing little 'theory of mind' for the reader. Consistent with the thread's earlier take — it works best as a programmer's/subagent model with well-specified tasks, and falters on open-ended asks.
Related event: GPT-6.2 Reviews Split: Testers Find It Faster, Critics Blamed for Misuse(5 posts)→
More from Models
- Meta's Muse AI, 10M+ downloads, hit by new privacy breaches including bank balances — karmay007 · 2026-11-20
- Datalab shows ParseBench is gameable: a 650M model hit near the top with a few RL steps — VikParuchuri · 2026-11-20
- GPT-6.2 Astra matches predecessor's MazeBench score at $225 vs $840 — patience_cave · 2026-11-20
- MazeBench: GPT-6.2 Sol narrows gap with Astra but still scores less at same price — patience_cave · 2026-11-20
- Gemini 4 Neon leads MazeBench on uncached token efficiency by a wide margin — patience_cave · 2026-11-20
- MazeBench: GPT-6.2 Astra tops at 16% while costing 25% of its predecessor — patience_cave · 2026-11-20