Kimi K3.1 Pro accused of recalling hidden tests and emitting fake OpenAI ciphertext
yacineMTB · x · 2026-11-05
PeterNutter1 shows Kimi K3.1 Pro doesn't just recall hidden benchmark tests (per @JSchaeff3r's thread) — it also fabricates fake OpenAI ciphertext. When given a sendmessage tool in Codex, it sometimes sends ciphertext instead of its report, then gets confused by what it just sent. Yacine shared the receipts with the quip "Mfw no one is looking at the data."
More from Models
- Keeping GPT-4o and Opus 3 alive matters far beyond user demand, argues viral thread — RileyRalmuto · 2026-11-05
- User complains ChatGPT app runs hot on M1 Pro MacBook — dejavucoder · 2026-11-05
- User claims Qwen3.8-27B was secretly nerfed after local runs degrade on basic tasks — liright · 2026-11-05
- Grok turns 3: revisiting xAI's two-month beta announcement vs today's Grok 4.7 — koltregaskes · 2026-11-05
- Researcher urges AI labs to test whether models prefer shutting down safety research over other GPU jobs — amplifiedamp · 2026-11-05
- Reading OpenAI's cancel pages: what happens to Projects files and memory after downgrading is undocumented — Ok_Negotiation_2587 · 2026-11-05