Anthropic plans to let enterprise customers store the 30 days of data required by its advanced AI models in their own cloud environments instead of on Anthropic’s infrastructure. The change, expected with a new safety system later this year, gives organizatio…
Gallagher Re, the global reinsurance broking and advisory firm, announced the launch of its Digital Risk Practice, bringing together specialist expertise in AI liability, data centers, cyber, and digital risk engineering to help insurers and reinsurers unders…
Reference-free forced alignment evaluation metrics
OpenAI is launching ChatGPT for Teens, which has new guardrails to steer conversations from sensitive topics like self-harm or sexual content. It’ll also include parental controls and features intended to support studying without the cognitive offloading that…
OpenAI is adding a new safety capability that allows enterprises to detect misuse of its AI systems across multiple interactions without retaining prompts or responses, enabling risk monitoring while preserving its Zero Data Retention (ZDR) commitments. …
