Fetching from the wire…
01
02
03
04
05
06
07
08
Public story · 2026-03-18 · source-backed
A Jones & Bergen study found GPT-4.5 fooled 73% of judges in five-minute Turing tests, but only when instructed to be laconic, use lowercase, make typos, and feign poor math. Without the dumbing-down prompt, pass rate dropped to 36%. The model succeeds by hiding competence, not demonstrating it. The Decoder
Each link below shares sources, entities, or timing with this story.
GPT competes with Claude / Shared entities / What happened next
Linked by a graph relationship (GPT competes with Claude); both cover GPT, Without; picks up the GPT thread on 2026-05-10.
GPT competes with Claude / Shared entity: GPT / What happened next / Tension
Linked by a graph relationship (GPT competes with Claude); both cover GPT; picks up the GPT thread on 2026-08-09.
Linked by a graph relationship (GPT competes with Claude); both cover GPT; picks up the GPT thread on 2026-08-06.
Linked by a graph relationship (GPT competes with Claude); both cover GPT; picks up the GPT thread on 2026-03-20.
Copilot uses GPT / Shared entity: GPT / What happened next
Linked by a graph relationship (Copilot uses GPT); both cover GPT; picks up the GPT thread on 2026-07-20.
Linked by a graph relationship (Copilot uses GPT); both cover GPT; picks up the GPT thread on 2026-05-13.
GPT competes with DeepSeek / Shared entity: GPT / What happened next
Linked by a graph relationship (GPT competes with DeepSeek); both cover GPT; picks up the GPT thread on 2026-06-19.
Copilot uses GPT / Shared entity: GPT / What happened next
Linked by a graph relationship (Copilot uses GPT); both cover GPT; picks up the GPT thread on 2026-05-11.