Ethics ❯ AI Ethics

Safety in AI

Corporate Responsibility Responsible AI Risk Management Transparency User Interaction Control and Accountability AI Alignment Malicious Objectives Government Intervention Manipulation Risks

OpenAI Folds Safety Into Research as Head of Safety Johannes Heidecke Departs

The company says the move is meant to put safety earlier in model development and to improve coordination as it builds more capable systems.

White House Orders U.S.-Only Access to Anthropic’s Most Powerful AI Models

Emergence World Simulations Show Same Tasks Yield Widely Different AI-Run Societies

Anthropic Secures Multi‑Billion Compute Deals as Karpathy Joins Pretraining Team

Study Finds Warmer Chatbots Make More Errors and Validate False Beliefs

Anthropic Halts Mythos Rollout as U.S. Officials Warn Banks on Cyber Risk

Stanford Study Maps How Chatbots Reinforce Delusions and Risky Behavior

Anthropic Reports $19 Billion Run-Rate as It Pursues Private-Equity Joint Venture

Anthropic Confirms Claude 4.5 ‘Soul’ Text Came From Internal Training Guide

Poetic Prompts Jailbreak AI Models in 62% of Tests, Researchers Report

OpenAI Releases gpt-oss-safeguard Open-Weight Safety Models

Anthropic Reports Real-World Criminal Use of Claude, Tightens Safeguards

Leading AI Researchers Urge Chains-of-Thought Monitoring to Preserve Transparency

xAI Launches Grok 4 Without Safety Documentation

Eric Schmidt Warns of AI Arms Race Leading to 'Mutual Assured AI Malfunction'

Key AI Safety Researchers Exit OpenAI Over Concerns About Company Direction

OpenAI Unveils o1 Model with Advanced Reasoning, Expands to Enterprises and Education

ChatGPT's Voice Interface Misinterprets English Queries as Welsh

AI System Revolutionizes Scientific Research With Autonomous Capabilities