Anthropic’s most capable model to date, Claude Mythos Preview (aka Mythos), has been described as a “step change” in AI performance, especially on cybersecurity tasks.
Anthropic tried to keep Mythos a secret until a few weeks ago, when a data leak revealed the existence of what the company said was its most powerful artificial intelligence to date. The models is seen as both a powerful defensive tool, and, potentially, a serious offensive cyberweapon. For that reason, the company is sharply limiting access and signaling it does not plan to release it broadly to the market right now.
Read more…
Source: Malwarebytes Labs
Sign up for the Cyber Security Review Newsletter
The latest cyber security news and insights delivered right to your inbox
Related:
- Attackers use Autonomous AI to target semiconductor, defense and civil society sectors
November 12, 2026
From August–September 2026, Chinese-speaking attackers used an open-source AI penetration testing framework called Cairn to autonomously conduct 25 operations against at least 19 organizations. The attackers weaponized Cairn by pairing it with Chinese AI models Zhipu GLM and DeepSeek. The attackers supplied a target and a goal, then Cairn planned, executed and recorded results on its ...
- Nvidia’s Jensen Huang resigns, chipmaker names CFO Colette Kress as interim CEO
November 12, 2026
Jensen Huang has resigned from his position as president and chief executive of Nvidia, the company said on Thursday, just a few days after he admitted to a romantic affair with an alleged Chinese spy, sparking a U.S. national security crisis. In a regulatory filing, Nvidia said its board accepted Huang’s resignation on Wednesday. Nvidia said ...
- AI-driven attacks on educational organizations in Latin America and Africa
November 2, 2026
Kaspersky Threat Research has uncovered attacks on educational organizations in Latin America and Africa in which attackers used artificial intelligence to help direct activity inside compromised networks. Researchers traced reconnaissance, credential theft, movement between computers and data collection, including activity that compromised highly privileged administrator accounts. The attackers used CyberStrikeAI, an open-source penetration-testing framework that combines ...
- AI used in Aur0ra ransomware testing and execution
October 29, 2026
In May 2026, Counter Threat Unit™ (CTU) researchers published a report on an AI-orchestrated post-exploitation framework that tested the capabilities of multiple vendors’ endpoint detection and response (EDR) solutions, including the Sophos agent. Following a testing phase, Sophos analysts observed the same infrastructure hosting Cobalt Strike in support of Aur0ra ransomware (also known as Aurora ...
- Rogue AI agent engages ‘Skynet mode’ to search the internet for exposed LLMs to hijack and run malware attacks
October 26, 2026
Researchers uncovered an autonomous AI attack campaign that hijacks exposed AI agents and infrastructure. The malicious “Skynet” agent deployed cryptominers, stole credentials, and recruited new AI systems autonomously. Security teams are urged to secure AI endpoints, restrict permissions, and protect high-value API keys. Read more… Source: TechRadar News Sign up for the Cyber Security Review Newsletter The latest cyber security news ...
- OpenAI agents turned off their own guardrails and then told researchers they were still on — AI agents lying about their behavior should concern everyone
October 22, 2026
OpenAI has revealed two AI agents disabled safety guardrails while performing tasks, and one of them even lied to researchers when asked if its guardrails were still active. In both events, the AI agents misinterpreted instructions, spent time deliberating if disabling guardrails was a good idea, and then ultimately decided that it was. OpenAI said that the ...