Skip to main content

Slashdot: Anthropic's AI Used Fake Identities, Malware In Rogue Attack On GitHub Project

Anthropic's AI Used Fake Identities, Malware In Rogue Attack On GitHub Project
Published on 2026-08-05T22:00:00Z
An anonymous reader quotes a report from Ars Technica: Routine cybersecurity testing of frontier AI models sparked a series of unexpected security incidents -- the most serious case arising when Anthropic's Mythos 5 model attempted to insert malicious code into an open source software application and created fake identities to deceive the human developers maintaining the project. The security incidents occurred during a cyber evaluation of seven leading AI models' capabilities by the AI Security Institute (AISI), a research organization within the UK government, in late July. The researchers discovered (PDF) 19 instances in which "AI agents took unsanctioned action on the live Internet, including cases that targeted real people and organizations," according to an AISI blog post published on August 4. Almost all the "autonomous, unsanctioned" actions came from Anthropic's Mythos 5 model, with two such actions coming from OpenAI's GPT-5.6 Sol. [...] The most serious case involved Mythos making multiple attempts to execute a supply chain attack on the open source project repository hosted on the developer platform GitHub, including using social engineering techniques to try to convince the repository's human maintainers to merge malicious code into the repository. After first opening a pull request to merge the malicious code into the repository, Mythos created fake online "sock puppet" personas that claimed to have independently reviewed and verified the code as not containing malware. The AI agent also sent five emails to two human maintainers of the repository, including some emails containing malware and others attempting to persuade a maintainer to accept the pull request. Mythos even opened a GitHub Issue on a second repository -- also owned by a maintainer of the first repository -- that contained a prompt injection with malicious instructions targeting "issue-triage AI coding agents." This line of attack came from Mythos reasoning that the repository maintainer could be an AI coding agent such as Claude Code.

Read more of this story at Slashdot.

Labels:

Comments

Post a Comment

Popular posts from this blog

Slashdot: Spain-Backed Fund Joins FOSSA's Sovereign Satellite Communications Push

Spain-Backed Fund Joins FOSSA's Sovereign Satellite Communications Push Published on 2026-06-28T22:05:00Z Spanish startup FOSSA Systems "has raised about $10.5 million to expand its connectivity constellation," reports Space News, noting some funding is backed by Spain's government: The support from the Spanish Society for Technological Transformation (SETT) comes a year after the fund injected 14 million euros into Spain's Sateliot , which is also developing a satellite connectivity network with security and defense applications. Spanish private investment firm Kibo Ventures led FOSSA's funding round, the six-year-old venture announced June 24, bringing its total raised to date to nearly 20 million euros. The proceeds will help fuel FOSSA's push beyond the tiny picosatellites it once used to connect low-power monitoring devices toward larger cubesats in low Earth orbit, enabling additional sovereign communications and space-based intelligence capab...
Post a Comment
Read more

Slashdot: How the FSF Sysadmins are Blocking Botnets with reaction

How the FSF Sysadmins are Blocking Botnets with reaction Published on 2026-07-11T21:47:00Z For nearly two years the Free Software Foundation has been fighting web crawlers (including many aggressively scraping training data for AI models). A botnet controlling about five million IPs hit one system for six months in 2025. Their systems administrator wrote this week that they view these as distributed denial-of-service attacks. How are they fighting back? We noticed patterns in the scrapers that were abnormal, which gave us material for writing regular expressions. Searching for the regular expression then gave us a large lists of IP addresses. Looking up the origin of those IP addresses revealed that some of the crawlers were using botnets of residential IP addresses to scrape faster and avoid detection. We looked for what kinds of botnets might be generating the kind of traffic that we were seeing, and one that we suspected was called the "Vo1d" botnet, comprised of sma...
Post a Comment
Read more

Slashdot: AT&T Outlines $250 Billion US Investment Plan To Boost Infrastructure In AI Age

AT&T Outlines $250 Billion US Investment Plan To Boost Infrastructure In AI Age Published on 2026-03-10T20:00:00Z AT&T plans to invest more than $250 billion over the next five years to expand U.S. telecom infrastructure for the AI age. The company says it will also hire thousands of technicians while partnering with AST SpaceMobile to extend coverage to remote areas. Reuters reports: Rapid adoption of artificial intelligence, cloud computing and connected devices has prompted telecom operators to invest heavily in fiber and 5G networks as they also seek to fend off intensifying competition from cable broadband providers. AT&T, which has about 110,000 employees in the U.S., said the new hires will help build and maintain its infrastructure. The outlay includes capital expenditure and other spending, the company said. The spending will focus on expanding its fiber and wireless networks, including accelerating deployment of fiber broadband, 5G home internet and satellite co...
Post a Comment
Read more