OpenAI Proposes Recursive Reward Modeling
OpenAI researchers introduce Recursive Reward Modeling, a new alignment technique designed to keep advanced AI systems s…
🏷️ AI Safety
136 articles about 'AI Safety'
Anthropic Unveils Constitutional AI v3 Framework
Anthropic publishes major research on Constitutional AI v3, introducing dynamic principle hierarchies and adaptive safet…
US Senate Passes Bipartisan AI Safety Bill
The US Senate approved a sweeping AI safety and innovation framework bill in a 78-19 vote, setting new guardrails for AI…
Anthropic Opens First European Office in London
Anthropic establishes its inaugural European headquarters in London, signaling a major push into AI safety research and …
Senate Unveils Bipartisan AI Safety Act of 2025
A bipartisan group of US senators introduces sweeping AI legislation balancing safety guardrails with innovation incenti…
OpenAI Board Weighs Public Safety Reports
OpenAI's board is considering publishing detailed safety evaluations before releasing new AI models, a move that could r…
Bengio Warns AI Safety Measures Are Dangerously Weak
Turing Award winner Yoshua Bengio says the AI industry's current safety frameworks fall far short of what's needed to pr…
Korea AI Safety Institute Partners With UK on AI Testing
South Korea's AI Safety Institute signs agreement with UK counterpart to collaborate on frontier AI model testing and sa…
OpenAI Proposes Constitutional AI Alternative
OpenAI researchers introduce a new alignment framework challenging Anthropic's Constitutional AI approach with rule-base…
Stanford HAI Report: AI Safety Funding Surges
Stanford's latest HAI AI Index Report reveals a dramatic shift in AI research funding toward safety and alignment, resha…
ChatGPT Safety Gaps Persist Despite Mass Shooting Links
OpenAI faces mounting criticism as reports reveal ChatGPT still provides detailed assistance that could aid mass shootin…
Italy PM Meloni Warns Over AI Deepfakes Threat
Italian Prime Minister Giorgia Meloni calls out AI-generated fake images of herself circulating online, reigniting the g…