Canada’s artificial intelligence minister said Tuesday the federal government now has access to Anthropic’s Mythos model — an AI platform so powerful that the company has withheld it from the general public.
AI Minister Evan Solomon said the government has signed onto Project Glasswing, which Anthropic launched earlier this year to allow the U.S. government and a limited number of top tech companies to use Mythos to scan its systems for security vulnerabilities.
Anthropic said earlier Tuesday it was expanding the program to 150 additional organizations in more than 15 countries.
“I can confirm that the Canadian government is part now of Project Glasswing, which allows companies to have access to Mythos,” Solomon told reporters after a cabinet meeting in Ottawa.
“So we do have access, we are part of Project Glasswing, I’ll confirm that, and through the Canadian Cyber Security Centre, we have access Mythos now. So it’s a very important step for Canadians and for our government to make sure that we have access and we can protect our institutions and our countries.”
Solomon said he couldn’t give details on which companies have access to Mythos.
Get daily National news
Get daily Canada news delivered to your inbox so you’ll never miss the day’s top stories.
Anthropic also declined to name specific companies getting access but said they cover industries such as health care, power, water, communications and hardware. The expansion also includes government organizations, the company said.
When it announced the creation of Mythos this spring, Anthropic said the model was so “strikingly capable” at hacking and cybersecurity work that it could only release it to a small group of trusted organizations.
The company said Project Glasswing was created in hopes of securing the world’s critical software from “severe” fallout that the new model could pose to public safety, national security and the economy — particularly if it was used in a major cyberattack.
Anthropic said Tuesday its partners using Mythos so far — including Apple, Google, Amazon and Microsoft — have found more than 10,000 security flaws they considered highly or critically severe, though some industry experts have said fears of unfettered hacking have been overstated.
Project Glasswing aims to bolster major software infrastructure before AI as powerful as Mythos is widely available from other providers, a prospect that Anthropic says is between six and 12 months away.
The company has said it aims to bring Mythos-class models to all of its customers, with additional safeguards, in the coming weeks.
—With files from the Associated Press and Reuters
© 2026 Global News, a division of Corus Entertainment Inc.