tech
OpenAI limits new AI models to 'trusted partners' at request of U.S. government
OpenAI said it previewed the new models' capabilities with the government ahead of the launch.
TL;DR
- OpenAI announced three new AI models: GPT-5.6 Sol, Terra, and Luna.
- The models are initially being rolled out to a small group of trusted partners at the request of the U.S. government.
- OpenAI previewed the models' capabilities and shared plans with the government before the launch.
- The company believes this short-term step is the best path to broader availability.
- GPT-5.6 Sol is described as OpenAI's strongest offering, with improvements in coding, biology, and cybersecurity.
- OpenAI stated that GPT-5.6 Sol is better at helping users fix vulnerabilities than carrying out attacks and does not cross its 'critical' cybersecurity risk threshold.