OpenAI updates AI behavior guidelines for controllability
OpenAI updated its Model Spec, restructuring AI behavior guidelines around five core values. New rules require GPT models to use conversation history to infer user intent and disclose assumptions. This aims to improve compliance and transparency in hard-to-control scenarios.
Summarized by AI from one outlet
Source
- THE DECODERRead original ↗
Latest
- Policy
OpenAI commits to regular Community Safety Reports to strengthen violence prevention
OpenAI announced that ChatGPT can automatically terminate conversations exhibiting violent fixation and committed to publishing semi-annual Community Safety Reports. The reports will detail progress in violence prevention and safeguard effectiveness. The first report is scheduled for early 2027, covering data from the second half of 2026.
- Policy
Malicious Claude Code Extension Hacks Cursor Marketplace
A hacker uploaded a malicious Claude Code extension to the Cursor marketplace, affecting up to 5,000 users by attempting to steal data and establish remote access. Cursor temporarily offline the marketplace and confirmed risk clearance, highlighting supply chain security risks in AI dev tools.
- Policy
OpenAI supports robots.txt for AI training preferences
OpenAI announced it will honor shared robots.txt signals, allowing website operators to communicate their AI training preferences to participating providers. This move aims to standardize how site owners manage how their content is used for model training, helping to clarify digital copyright and training data boundaries.
- Policy
Report: AI compute expansion could raise US power carbon emissions 42% in five years
A Sierra Club analysis suggests that building fossil fuel infrastructure to meet AI data center electricity demands could increase US power industry CO2 emissions by 42% in 2030 compared to 2025. This would offset 15 years of greenhouse gas reduction efforts, raising concerns about environmental impact and regulatory policy.
- Policy
Wikimedia warns AI copyright ruling could threaten open web
The Wikimedia Foundation warned that binary AI copyright rulings could threaten the open web. It advocates protecting LLM training on copyrighted works while restricting chatbots from regurgitating content.
All news → · All rights belong to the original creators