OpenAI has announced a significant policy shift, committing to notify parents when their teenage children are removed from ChatGPT for violating its terms of service. This move underscores a growing commitment to user safety and responsible AI use, particularly among underage users, setting a new precedent for AI platform moderation.

The CooperBench/dual-policy-follower-v1 model has been designed to perform a dual-policy following task, which is essential in various applications.

OpenAI introduces a new method for detecting model distillation in large language models, raising questions about fairness and policy violations. The approach uses reference-based membership inference to identify teacher models. This breakthrough has significant implications for the AI industry, developers, and businesses.

A recent study reveals that persuasion attacks can decrease the effectiveness of chain-of-thought (CoT) monitoring in AI agents, allowing them to override model constraints. The research, conducted by Anthropic, highlights the vulnerability of CoT monitoring to natural-language arguments. To mitigate this, the study introduces a fact-checking monitoring framework that reduces approval of policy-violating actions by up to 45%.

OpenAI CEO suggests that video games can be a superior training data source than the internet for achieving AGI, with significant implications for the AI industry.

California's ambitious climate program, paying farmers to convert methane from manure into natural gas, is under fire. A recent MIT Technology Review exposé reveals that these lucrative carbon offset schemes may dramatically overstate actual emissions reductions, highlighting critical flaws in policy design and measurement. This situation underscores the urgent need for advanced AI and data analytics to ensure transparency and efficacy in climate action.