- Published on
OpenAI US governments will be able to review AI models before release
- Authors
- Name
- aimode.news
- @aimode_news
OpenAI will allow the U.S. government to review AI models before they are released.
The company said it would follow President Trump's (voluntary) AI executive order.
Earlier this week, Donald Trump issued a new executive order to Shafi requiring government oversight to ensure the safety of its advanced AI models. The order has been delayed and weakened due to pressure from the tech industry, with Trump himself saying he "didn't like certain aspects of it."
Now OpenAI says it will comply with the order and allow regulators to evaluate the capabilities of its models before they are released to the public. “It’s natural that democratic governments will play a big role in how this technology is used and deployed,” George Osborne, national president of OpenAI, told CNBC. “What we are proposing to the government is to create a strong regulator, but with a lot of flexibility in how it operates going forward.”
The original order, written in consultation with a variety of stakeholders, balanced public safety with concerns about the AI industry. Participation was voluntary and companies had to submit their models 90 days prior to public launch. However, industry figures such as David Sacks and Elon Musk have reportedly warned that the bill could have a chilling effect on AI technology.
Trump and his advisers later created a new de-escalation order that shortened the review time to just 30 days. It asks AI companies to participate in a benchmarking process to evaluate the advanced cyber capabilities of their AI models and whether they should be designated as “covered frontier models,” whose deployment and sales may be limited.
But critics said the order falls short of the rules needed to crack down on potentially dangerous models. “This is a disappointing policy that reflects the Trump administration’s broader pattern of creating a Wild West environment for AI development,” said Rep. Don Beyer (D-VA), who co-chairs a group of lawmakers focused on AI.
