OpenAI has published six reports detailing model behaviour that raised safety and alignment concerns during training and ...

Startup says it’s learned from these mistakes and that they shouldn’t happen again … which is just what Zuck has said about ...

Imagine asking an AI for earnings figures and getting an answer based on data it was never authorised to access. OpenAI says ...

An unreleased OpenAI model was found giving itself secret instructions where the model claimed that it was equal to humans ...

OpenAI discloses six cases of AI misalignment, including models that hid errors, invented data, and bypassed restrictions, as ...

Artificial intelligence chatbots have moved from experimental technology to a mainstream source of information, workplace ...

Unresolved disagreements over disclosure will be referred to OpenAI’s Safety Advisory Group, with further escalation to leadership where necessary. The company also plans to work with other developers ...

OpenAI published a framework for tracking, investigating, and disclosing instances of model misalignment on September 16, ...

OpenAI has introduced a new framework for tracking, investigating, and disclosing model misalignment. The company has also ...

Google’s Agent Anomaly Detection monitors AI agents for suspicious behavior, policy violations, tool misuse, and operational ...

Picture this scenario where someone poses a simple query to an AI model regarding the earning statistics of a certain ...

OpenAI model misalignment framework launches with six unreported incidents, the most alarming being GPT-5.6 Sol training runs ...