Claims Browser

Society faces a critical challenge in whether it can slow, test, and govern increasingly capable AI systems before ordinary incentives reward models that are smarter, less reliable, and harder to contain

"The stakes are whether society can slow, test, and govern increasingly capable systems before ordinary incentives reward models that are smarter, less reliable, and harder to contain"

Current AI tools have specific use cases where they are actually good, but they can also distort judgment in certain applications

"discuss what current AI tools are actually good for, where they distort judgment"

Writing still matters as a way of thinking even in the age of AI

"why writing still matters as a way of thinking"

OpenAI has data showing how people use ChatGPT worldwide at the country level

"New OpenAI Signals data shows how people use ChatGPT worldwide, with country-level insights on adoption, usage trends, and evolving behavior."

OpenAI and APA are developing evidence-based guidance and safeguards for responsible AI use in youth mental health

"OpenAI and the American Psychological Association advance evidence-based guidance, resources, and safeguards for responsible AI use and youth mental health"

ChatGPT has introduced an improved GPT-5.6 Sol model with better accuracy and consistency

"ChatGPT introduces improved GPT-5.6 Sol with better accuracy and consistency"

Free users now have expanded access to ChatGPT features

"expanded access for free users"

GPT-5.6 Luna offers unlimited everyday chats

"unlimited everyday chats with GPT-5.6 Luna"

OpenAI is conducting preliminary cybersecurity evaluations for Astra

"OpenAI is sharing preliminary cybersecurity evaluations for Astra"

Goodfire has built Silico, a research platform priced at $1,000 per month

"Silico, the $1,000-per-month research platform Goodfire built for itself"

Fine-tuning and reinforcement learning often amplify behaviors that are already latent in pre-training

"fine-tuning and RL often amplify behaviors already latent in pre-training"

Interpretability techniques can identify the data and features driving unwanted model updates

"interpretability can identify the data and features driving unwanted updates"

Models do not store concepts as simple one-hot features, but as sparse mixtures of meaningful subspaces

"models do not store concepts as simple one-hot features, but as sparse mixtures of meaningful subspaces whose geometry determines what kinds of steering and control work"

The geometry of concept manifolds determines what kinds of steering and control techniques will work on models

"sparse mixtures of meaningful subspaces whose geometry determines what kinds of steering and control work"

Modern interpretability research may be moving beyond its reputation as a toy-model science

"modern interpretability may be moving beyond its reputation as a toy-model science"

Steering techniques can fail when applied off-manifold from the concept subspaces

"why steering can fail off-manifold"

ChatGPT Business now offers premium seats with 5x more usage capacity compared to standard seats

"Premium seats are now available on ChatGPT Business, with 5x more usage"

GPT-5.6-Cyber is available through Daybreak Red for authorized vulnerability research, exploit validation, and security testing

"available through Daybreak Red for authorized vulnerability research, exploit validation, and security testing"

Model ML uses GPT-5.6 Sol to perform finance work including research, analysis, and generation of editable PowerPoint decks and Excel workbooks

"Model ML uses GPT-5.6 Sol to carry finance work from research and analysis through editable, traceable PowerPoint decks and Excel workbooks"