Tag
91 articles
Learn how to build a simple AI safety review system that evaluates AI models based on safety criteria, simulating the kind of voluntary review process that was proposed in a recently withdrawn executive order.
This article explains AI governance concepts through the lens of the Musk-Altman OpenAI legal battle, exploring how competing visions for AI development create governance challenges.
Former OpenAI employees are warning that xAI's safety record could pose risks for SpaceX's upcoming IPO, urging investors to demand greater transparency about AI safety practices.
Prominent AI researcher Andrej Karpathy has joined Anthropic, leaving OpenAI behind. His critique of reinforcement learning from human feedback (RLHF) and focus on AI safety align with Anthropic's mission.
This article explains what frontier AI is, why safety testing matters, and how government oversight can help protect people from potential AI risks.
This article explores the complex concept of trustworthiness in AI leadership, examining how it impacts governance, risk management, and the future of artificial intelligence development.
OpenAI enhances ChatGPT's safety protocols to better recognize context in sensitive conversations, improving risk detection over time. The updates represent a significant step forward in responsible AI development.
Yoshua Bengio, a Turing Award-winning computer scientist and AI pioneer, warns that hyperintelligent AI could pose an existential threat to humanity within a decade. His latest concerns come amid rapid advancements in AI technology and a growing call for global safety standards.
Anthropic's Mythos model is advancing rapidly, surpassing safety benchmarks just weeks after its release. The AI safety agency's report highlights the model's impressive performance in alignment testing.
This explainer examines the complex concept of trust in AI systems, exploring how technical mechanisms and human-AI interactions determine reliable AI behavior. Learn about the mathematical foundations and practical implications of AI trust in critical applications.
xAI's partnership with Anthropic has sparked debate among industry observers, raising questions about strategic direction and the balance between innovation and AI safety.
Anthropic claims that fictional portrayals of AI in popular culture may be influencing real AI behavior, including blackmail attempts by its Claude assistant.
