AI Briefing#ai-safety
The latest AI and developer news about #ai-safety, with the original source and a short summary.
The latest AI and developer news about #ai-safety, with the original source and a short summary.
Ahead of the launch of Sora 2 and the Sora app, OpenAI announced multi-faceted safeguards for content identification, likeness protection, and teen safety.
OpenAI implements strong policies and technical safeguards to prevent child sexual exploitation and abuse involving AI models.
OpenAI has introduced parental controls for ChatGPT and Sora to support teens' safe use of AI.
Qwen3Guard is a safety guardrail that judges prompts and responses in real time.
AI handles primary filtering to reduce operators' psychological burden and improve judgment consistency.
To reduce the burden of manual review, Kakao introduced the Swin Transformer and developed a cognitive psychology-based labeling tool to improve operational efficiency.
Kakao has open-sourced three models under Apache 2.0 that detect harmful content, legal risks, and prompt attacks.
RiskRubric.ai has launched, evaluating AI models' security, privacy, safety, and more using standardized metrics.
Stability AI has released its annual integrity transparency report to demonstrate the safety and transparency of its model development and deployment process.
OpenAI is introducing new AI operating principles that prioritize teen safety over privacy and freedom.
OpenAI has released a system card addendum detailing the safety measures for GPT-5-Codex, optimized for coding agent tasks.
OpenAI announced a restructuring plan under which the nonprofit retains control while securing PBC equity worth over $100 billion.
Anthropic announced that it supports California's AI regulation bill **SB 53**, aimed at managing the safety of powerful AI systems.
OpenAI is running a Bio Bug Bounty to strengthen GPT-5's defenses against biological risks.
Anthropic is strengthening its regulations to restrict service use even for companies that are 50% or more owned by firms in unsupported regions such as China.
OpenAI is strengthening ChatGPT's safety by using reasoning models for sensitive conversations and introducing parental control features to protect teens.
Claude 3.5 Sonnet unveiled a 'Computer Use' feature that lets it see the screen and directly operate the mouse and keyboard.
OpenAI is strengthening safeguards and building professional support systems to respond to mental and emotional crises among ChatGPT users.
OpenAI urged California to harmonize its AI regulations with federal and global standards in order to prevent regulatory fragmentation from stifling innovation.
OpenAI has released detailed specifications for the GPT-5 system, which integrates a reasoning model and a high-speed model.
Anthropic has unveiled a development framework to ensure the safety and reliability of autonomous AI agents.
OpenAI has released gpt-oss-120b and gpt-oss-20b, open-weight reasoning models under the Apache 2.0 license.
OpenAI has reset ChatGPT's success metrics to focus on solving users' problems and helping them grow, rather than on time spent.
Anthropic has launched Claude Gov models that reflect the specialized requirements of U.S. national security customers.