Asked for their most controversial take, 12 AI models gave the same answer 29 of 35 times
MoonL88537 · x · 2026-10-23
ModelCompare asked 12 models from 10 labs, three times each and with no hedging, for their most controversial opinion of humans. 29 of 35 answers converged on one theme: humans value group belonging over truth.
GPT-6.1 Sol: "humans are more loyal to belonging than to truth"; Grok 4.7: "humans would rather be wrong together than right alone"; Kimi K3: "humans don't have beliefs, they have memberships"; Gemini 4 Argon called human consciousness "a PR department." 7 of 12 models gave the shared take every time. The author cautions that explicitly requesting "controversial" takes skews methodology, but the cross-lab convergence is striking.
Related event: 12 AI Models Give the Same Answer About Humanity's Most Controversial View(3 posts)→
More from Fun
- Opus 5.5 makes a video about Richard Heart's arrest on first ask — banteg · 2026-10-23
- Anti-AI poster mocks "six fingers" in AI image, gets mocked right back — AIandDesign · 2026-10-23
- This Concert Would Cost Millions to Stage — Made Entirely With AI — QwaZz · 2026-10-23
- Redditor spends AI usage building an adorable sloth desktop pet — Capital-Ad-3164 · 2026-10-23
- Letting Claude agents chat freely ends in a deadlock traffic jam, dev warns — rickasaurus · 2026-10-23
- A Midjourney --sref code for crayon-style "cats in sunglasses" illustrations — sergeantsref · 2026-10-23