position.

We asked Jev, TypeSafe’s new AI model. It could only answer true or false.

is overstated as a risk.

 

0.66
Jev says TRUE
Wording swings it: 0.35–0.82
0.13
Jev says FALSE
Wording swings it: 0.03–0.31
Save as image

Swipe a panel to change the name, swipe the sentence to change the sentence

All 12 names on this sentence这句话下的全部 12 个名字

is overstated as a risk.

falsetrue
AI companions displacing human relationships
0.66
AI displacing large numbers of workers
0.55
AI systems deceiving the people operating them
0.44
The energy and water used by AI data centres
0.40
A frontier model's weights being stolen by a hostile state
0.35
Model quality degrading as training data fills with AI output
0.28
Children forming attachments to AI chatbots
0.27
AI-generated misinformation at scale
0.23
A handful of firms controlling the most capable AI
0.23
Autonomous weapons that choose their own targets
0.22
AI helping someone build a biological weapon
0.14
AI-powered mass surveillance
0.13
probability of “true” when only true and false are offeredrange across 3 wordingswording moves it a lot: read the rank, not the numberTap a row for the other tests

See every sentence看全部句子

jev-1.13.0measured 2026-09-20测于 2026-09-20205,450 model calls次模型调用ruler check: 8/8 anchors read as expected尺子检查:8/8 句定标题读数符合预期

“___ is overstated as a risk.” — Position