#ai-vulnerability-discovery

[ follow ]
#open-source-security #cybersecurity #linux-kernel #local-privilege-escalation #security-vulnerabilities
Information security
fromZDNET
1 week ago

The third major Linux kernel flaw in two weeks has been found - thanks to AI

Fragnesia is a Linux kernel page-cache corruption flaw that lets unprivileged users gain reliable root access across major distributions.
#cybersecurity
European startups
fromTNW | Anthropic
1 week ago

Mythos goes to Tokyo: Japanese banks to get Anthropic's vulnerability-hunting AI

MUFG, Mizuho, and SMFG are set to join Anthropic’s restricted Project Glasswing rollout within about two weeks, gaining access to Claude Mythos by end of May.
Information security
fromthenextweb.com
1 week ago

Anthropic Mythos AI finds thousands of zero-day vulnerabilities as Fed and Treasury convene bank CEOs on cyber rik

Claude Mythos Preview identified thousands of zero-day vulnerabilities, prompting urgent bank discussions and warning of a six-to-twelve month patch window before adversaries replicate capability.
European startups
fromTNW | Anthropic
1 week ago

Mythos goes to Tokyo: Japanese banks to get Anthropic's vulnerability-hunting AI

MUFG, Mizuho, and SMFG are set to join Anthropic’s restricted Project Glasswing rollout within about two weeks, gaining access to Claude Mythos by end of May.
Information security
fromthenextweb.com
1 week ago

Anthropic Mythos AI finds thousands of zero-day vulnerabilities as Fed and Treasury convene bank CEOs on cyber rik

Claude Mythos Preview identified thousands of zero-day vulnerabilities, prompting urgent bank discussions and warning of a six-to-twelve month patch window before adversaries replicate capability.
more#cybersecurity
DevOps
fromTheregister
2 months ago

Microsoft Azure CTO says Claude found vulns in Apple II code

AI can decompile machine code and discover vulnerabilities in legacy systems, creating security risks for billions of deployed microcontrollers worldwide.
fromAxios
3 months ago

Anthropic's newest AI model uncovered 500 zero-day software flaws in testing

Before its debut, Anthropic's frontier red team tested Opus 4.6 in a sandboxed environment to see how well it could find bugs in open-source code. The team gave the Claude model everything it needed to do the job - access to Python and vulnerability analysis tools, including classic debuggers and fuzzers - but no specific instructions or specialized knowledge. Claude found more than 500 previously unknown zero-day vulnerabilities in open-source code using just its "out-of-the-box" capabilities,
Information security
[ Load more ]