Devs complain Claude is 'practically blind' at image understanding, wildly hit-or-miss
MoonL88537 · x · 2026-10-20
Developers vent that Claude's image understanding is 'practically blind' and wildly hit-or-miss — 'are we looking at the same thing?' They also reference community claims that Claude doesn't truly look at images even when provided in context.
More from Fun
- Yacine on his moat: no shame means doing the 'dumb' thing that works — yacineMTB · 2026-10-20
- Researcher jokes AI field is 'rudderless' after a year without Altman's updated forecasts — generativist · 2026-10-20
- Ask Claude how long a feature takes: 'three weeks' means it'll be done in 30 minutes — dreamwieber · 2026-10-20
- Mumbai now world's 3rd most polluted city, memed as the ultimate RL 'pollutionBench' — dejavucoder · 2026-10-20
- Geek writes a chord progression on a graphing calculator, continuing math-driven drum & bass experiments — measure_plan · 2026-10-20
- One of the biggest tells of AI writing is being scrupulous about evidence — generativist · 2026-10-20