Home · Technology · Sep 28 archive
OpenAI to Roll Out New AI Model With Enhanced Safety Measures
Developing
In Short: OpenAI will begin rolling out its newest AI model, Astra, to select customers with added safety features.
OpenAI announced Thursday it will start rolling out its newest and most powerful artificial intelligence (AI) model, GPT-6, also known as Astra, to select customers. The San Francisco-based company said it had implemented stronger safeguards to address security concerns.
OpenAI President Greg Brockman told reporters that safety is now the top priority at this level of AI capability. The company paused some model development for two weeks this summer following a security breach at AI platform Hugging Face involving two models it was testing.
Despite the breach, Astra itself was not involved in the hack. OpenAI said it developed Astra with enhanced safeguards. Some cybersecurity customers will gain access to the new model Thursday, with a wider rollout to follow.
The US state of Alabama has opened an investigation into OpenAI over the Hugging Face breach, citing the company's lack of oversight and safeguards. Concerns about advanced models have grown since then, following incidents involving systems from OpenAI and rival Anthropic.
In a blog post, OpenAI said Astra can handle a wide range of tasks, including website creation, scientific analysis, game development, cybersecurity, and coding. The company highlighted the model's ability to significantly reduce the time spent on tasks such as apartment hunting, cutting it from six hours to under 10 minutes.
OpenAI previously had an agreement with Microsoft, one of its earliest and largest investors, under which an exclusivity clause would end once OpenAI reached AGI.
What's still developing
- Just over a year has passed since OpenAI launched the previous version of its flagship model, GPT-5.