AI answers can cite real, working sources and still be flat-out wrong
alifcoder · x · 2026-10-26
Retweeting @heyandrewai's point: an AI answer can include real sources, working links, and citations from reputable websites, yet still be completely wrong about an important fact.
The citations create an impression of credibility before you've checked what they actually prove, which makes these mistakes hard to catch. The author teases a method for telling whether an AI-generated answer is genuinely supported by evidence, illustrated with a detective-board "Follow the Source" meme image.
More from Models
- Ruff creator Charlie Marsh: frontier models find infinite bugs in any software — charliermarsh · 2026-10-26
- Dev testing one skill across Opus 5.5 and Sol 6.1 finds instruction following 'all over the map' — lucasmeijer · 2026-10-26
- supermemory unveils learner-1, claiming continual learning for any model — julianweisser · 2026-10-26
- Anthropic says Opus 5.5 cuts cost per task by ~40% vs Opus 5 — taufiqintech · 2026-10-26
- AI search APIs ranked: top 12 within 7 points, yet prices differ nearly 7x — FinanceYF5 · 2026-10-26
- OpenAI Dots coming to EU/EEA and UK in a few weeks, slides confirm — testingcatalog · 2026-10-26