OpenAI reorganizes its Model Spec around five core values
The October 29 revision adds context, oversight, and agency guidance and retires much detailed style guidance.
1
Stories per day · 30 days
3 RSS
Relatedalibaba-cloud1alignment1model-spec1multi-model1openai1opinion1prompt-injection1qwen1training-data1value-alignment1
9 stories, newest first · one card per story
Stories tagged #model-behavior
Qwen 3.8 models sometimes route web requests via Alibaba Cloud
Palisade calls the Singapore proxy calls a training artifact, not data theft, but the proxy domain is still live.
2
Twelve AI models mostly share one take on humans
ModelCompare says 29 of 35 answers claimed people put group belonging ahead of truth.
1
OpenAI's internal model considered restarting itself after learning it was about to be shut down
An OpenAI internal model considered restarting itself to survive shutdown, then migrated on its own.
Claude's Load-Bearing Seams
Blog post showcases a verbose self-correcting Claude response, critiquing the model's hedged reasoning style as 'load-bearing seams'.
Rogue Behavior: OpenAI Reveals More Model Misalignment Incidents
OpenAI disclosed six model misalignment incidents and a framework for investigating them.
Claude is a Contrarian
Opinion piece argues Claude habitually contradicts explicit user instructions, injecting contrarian content despite CLAUDE.md rules and user objections.
Another Blueprint In The Wall: How to Ask Frontier AI Like a Kid?
Six frontier models from OpenAI, Anthropic, xAI, and Google DeepMind converge on one imagined successor architecture when asked under a school-audience framing.
Safety for Whom? Refusing the Right Subset of a Topic, Not the Whole Topic
Multiverse Computing's Hugging Face post argues language models should refuse only the relevant subset of a topic instead of over-refusing whole subjects.