Model updates

Official news, first rumor reports and substantive progress. Each entry keeps the information published at that time.

• Artemis Security launches Orion-1, a cyber-defense model that it says beats Claude Opus 5.5 and GPT-6 Sol on its own benchmark, per The Decoder

  US startup Artemis Security launched Orion-1, a model that investigates and responds to attacks on corporate networks by linking log signals from auth services, cloud setups, endpoints, SaaS apps and network gear to a single attacker, then assigning a confidence score and recommending or starting a response. On Artemis's own Decision-Grade Readiness benchmark it scored 80.8 on attack reconstruction versus 58.3 for the best general model tested, GPT-6 Sol, in a comparison that also included Claude Opus 5.5, Grok 4.7 and Kimi K3. Artemis says Orion-1 was fine-tuned on tasks from its own security operations without customer data, but does not say whether it is built on another provider's model. It is in private preview for select customers, with high-impact actions requiring human approval by default.

  Reported by The DecoderPublished newsOct 14Source: The DecoderPart of Orion-1

RSS for this view →