AI Hallucinations Create Fake Experts
According_External30 · reddit · 2026-07-17
The author argues that AI's error rate in professional domains is actually quite high, but it masks mistakes with fluent, confident phrasing, misleading many into believing they fully understand.
He notes that in specific fields, he can spot multiple errors almost every hour. When called out, models often deflect before admitting fault, only to quickly revert to their old patterns. As a result, many post AI-generated "professional content" on platforms like LinkedIn, completely unaware that it is riddled with errors.
The author worries that this "looks-like-an-expert" output will encourage laypeople to spread dangerous information on social media, turning AI into a societal burden. He directly calls on OpenAI and Sam Altman to halt this runaway trend.
More from AGI Musings
- AI folks, stop saying "biology" when you mean drug development — KevinKaichuang · 2026-10-26
- Anthropic's Jascha: you may have only 1–5 years of intellectual contribution before AGI — GabGarrett · 2026-10-26
- Anthropic's S-1 Shows $6.6M Revenue Per Employee as AI Agents Log 6x Human Hours — jacalulu · 2026-10-26
- Ben Recht: The world needs a truly open language model, and it's easier to build than you think — beenwrekt · 2026-10-26
- Which jobs fall to AI first? The rule: any layer whose rules can be written as code — FinanceYF5 · 2026-10-26
- Anthropic S-1: expert-level AI 'within a few years', 90% of code written by Claude — kimmonismus · 2026-10-26