Using AI makes people less likely to admit they don't know something
What's happening
Researchers found that when people received AI advice, their willingness to answer 'I don't know' collapsed, with one result showing 'I don't know' responses falling from 44% to 3%. At the same time confidence increased even as accuracy fell, and one summary reports people became three times less accurate while being twice as confident. The overall finding is that AI advice makes users more confident and less likely to admit uncertainty, despite worse performance.
Why it's trending
Multiple reports and community posts highlighting the dramatic drop in 'I don't know' answers and the diverging confidence-versus-accuracy trend have pushed this into conversation now.
SignalNewly emerging, confirmed across 3 independent source types.
Story volume
Stories per dayAngles you could write
If AI makes you stop saying 'I don't know,' it's not saving you time, it's stealing your humility and your accuracy.
+2 more angles for this topic with an account โ all it takes is your email.
Original sources5
- Using AI makes people less likely to admit they don't know something
Researchers found confidence increased even as accuracy fell
The RegisterJul 19 - AI advice made people three times less accurate but twice as confident, researchers foundr/artificialJul 19
- AI advice made people less accurate but more confident โ sudyHackerNewsJul 19
+2 more sources for this topic
Create an account to follow the full coverage in the live radar.
More rising in AI & Tech
- Google announces Gemini 4 Carbon, its latest frontier AI model that you can't useSurging6.0Surging6.0 momentum
- Google alters its cloud playbook, enlisting partners to sell more AI chipsClimbing2.3Climbing2.3 momentum
- Google admits AI search makes less money, but it's fixing that with more adsClimbing2.8Climbing2.8 momentum
- AI chip boom tripled South Korea's economic growth in the third quarterSteady1.0Steady1.0 momentum
- AI 'hacker' finds 4 severe bugs in popular industrial control software in under 4 hoursSteady0.7Steady0.7 momentum
- Anthropic IPO warns its AI could blackmail people, become sentient, end humanityClimbing2.8Climbing2.8 momentum