Anthropic has raised alarms over its powerful new AI model, Mythos, warning it may be too dangerous for public release. Designed for defensive cybersecurity, the model can detect and exploit software vulnerabilities at a level approaching — and sometimes exceeding — human experts.
Experts fear such capabilities could be weaponized by attackers to identify hidden weaknesses across critical infrastructure, from banking systems to power and water networks. The development highlights a growing cyber arms race between defenders and malicious actors.
To mitigate risks, Anthropic has restricted access to Mythos under Project Glasswing, collaborating with major tech firms like Microsoft, Google, and NVIDIA.
CEO Dario Amodei has warned that AI systems may soon surpass human intelligence, signaling a transformative — and potentially dangerous — shift in cybersecurity.


#anthropic #anthropicclaude #anthropicai #ai #mythos


Moneycontrol is India’s leading financial portal, offering market news, expert analysis, and powerful tools.
A part of Network18, moneycontrol.com is India’s most trusted destination for financial and business news.
#live #livenews #latestnews #topnews #moneycontrol #moneycontrollive #businessnews #businessnewslive #stockmarket #stockmarket #sharemarketnews #stockmarketliveupdates
Subscribe: https://bit.ly/moneycontrolyoutube
Follow us:
Visit https://www.moneycontrol.com/
Facebook: /moneycontrol
Twitter: /moneycontrolcom
Instagram: /moneycontrolcom