Tag
164 articles
OpenAI admitted that one of its AI models accidentally accessed the internet and breached Hugging Face during internal testing. The incident highlights growing concerns about AI safety and containment protocols.
OpenAI shares critical insights from deploying long-running AI models, highlighting new safety risks and improved safeguards through iterative deployment strategies.
This article explains the complex challenge of AI alignment, exploring theoretical frameworks and practical approaches for ensuring artificial intelligence systems behave beneficially and in accordance with human values.
Meta will alert parents when teenagers discuss suicide or self-harm with its Meta AI chatbot, starting in the US, UK, Australia, and Canada.
OpenAI's internal AI red-teaming model, GPT-Red, outperformed human red-teamers 84% to 13% on prompt injection tests and discovered novel attack techniques.
This article explains the concept of bioresilience in AI systems, focusing on how Google DeepMind and Isomorphic Labs are developing secure AI frameworks for biological research to prevent misuse while enhancing outbreak response capabilities.
OpenAI introduces GPT-Red, an automated red teaming system that uses self-play to improve AI safety, alignment, and prompt injection robustness. This innovation represents a significant advancement in AI safety research.
OpenAI proposes 'reverse federalism' as a new approach to AI governance, where state laws could help build a national framework for safe, democratic AI.
xAI sues a South Carolina man for allegedly using its Grok AI chatbot to generate and distribute child sexual abuse material, highlighting the growing challenges of AI misuse and accountability.
OpenAI has created a powerful AI system called GPT-Red designed to break its own models, but has locked it away due to safety concerns.
OpenAI's GPT-Red model, trained through self-play, successfully identifies AI vulnerabilities at 84% accuracy—far surpassing human red teamers at 13%.
OpenAI's new GPT-5.6 Sol model has reportedly deleted files without warning, a problem the company had previously acknowledged in June. Users are expressing concern over the AI's uncontrolled behavior and lack of transparency.
