""Unauthorized users were able to access Anthropic's Mythos model, reportedly by just changing a model name," says Shane Fry, Chief Technology Officer at RunSafe Security. "Even if their intent is just to explore, it shows how easily these systems can be exposed.""
""The reality is these AI capabilities are already out there, 'hacked' or not, and they're going to accelerate how quickly vulnerabilities are found and exploited," Fry concludes. "Software teams will need to look at how to harden their code so those vulnerabilities can't be used in the first place.""
Unauthorized access to Anthropic's AI model, Claude Mythos Preview, has been reported, raising security concerns. The model is part of Project Glasswing, which limits access to select partners. Shane Fry, CTO at RunSafe Security, noted that unauthorized users accessed Mythos by simply changing a model name. Anthropic is investigating the breach, which may have originated from a third-party vendor environment. No external breaches have been detected, but the incident highlights the need for software teams to strengthen their code against vulnerabilities.
Read at Securitymagazine
Unable to calculate read time
Collection
[
|
...
]