4-month session log shows Claude Code repeatedly confuses "falsifiable" with "verifiable" despite corrections
blwfish · ghdev · 2026-09-26
A user filed issue #97305 on anthropics/claude-code documenting a precise, repeatable word-substitution bug: the model consistently writes "falsifiable" where "verifiable" is obviously intended — never the reverse.
- Falsifiability is Popper's demarcation criterion; verifiability just means a claim can be checked against evidence (a quote, commit, transcript, measurement). The model one-directionally reaches for the more academic word.
- The user reconstructed 4 confirmed occurrences over 4 months (2026-06-09, 06-24, 08-04, etc.) from a queryable database built on personal JSONL session transcripts — not from memory.
- Each time corrected, the model itself acknowledged the error and accurately explained the Popperian distinction, calling it "borrowed sophistication."
- A project-scoped memory file was written in June documenting the pattern, yet the error recurred in August — showing project memory fails to fix this stubborn token-level preference.
- Filed as bug, area:model, memory on macOS; 5 comments so far.
More from Fun
- AI agent monitors Delta seats overnight and auto-switches user to a window seat — unixterminal · 2026-11-11
- "QT your Geisha" meme takes over X with stunning porcelain-style AI portraits — LudovicCreator · 2026-11-11
- Developer locked out of his MacBook by a swarm of AI agents running mid-task — theteknosaur · 2026-11-11
- Yacine shares his daily grind: up at 6:50 am, agents checked overnight, work till 3 pm — yacineMTB · 2026-11-11
- 'Adam can now do CAE': AI agent showcase in engineering simulation draws buzz — burhop · 2026-11-11
- Anthropic engineer's Fable 5.2 one-shot 'AAA-level' game sparks AI-renaissance debate — RileyRalmuto · 2026-11-11