AI Computer Use Models Mocked for Getting Lost After UI Updates
menhguin · x · 2026-08-09
User @menhguin humorously criticized current AI 'computer use' capabilities, joking that the models seem to be trained exclusively on data of 'clicking update and then trying to find the same button you used on Cursor just last week, while ignoring literally everything else added in the past year.' This highlights the struggles agents face with dynamic UIs and software updates.
More from coding & agent
- Dev Complains About Poor Browser Use Experience in Coding Agents — MengdiWang10 · 2026-08-09
- Atlassian designer ships 125 PRs in 6 months using AI coding tools — davidhoang · 2026-08-09
- “The agent is working in the sandbox”: Usually implies a spectacular fail — haydendevs · 2026-08-09
- Vibe Coding Too Pricey: Dev Burns 1.5M Tokens in Minutes via Codex CLI — afl_ext · 2026-08-09
- AI Code Review Pitfall: Vague Prompts Disguise Core QA Pipeline Issues — ClickOk5811 · 2026-08-09
- Liquid AI's New Model Outperforms 4x Larger Peers in Private Benchmarks — JosephJacks_ · 2026-08-09