Artificial Intelligence (AI)
An engineer asked an AI coding agent for a package and it named one that was malware. Only a company rule to review any AI-recommended code on GitHub first caught the slop-squatting attempt before install.
Industrial Control Systems (ICS)
Five US agencies, led by NSA, FBI and CISA, warn that an active campaign is using AI-assisted development to build tools targeting internet-exposed Siemens S7 Series PLCs, in what may be the first US-agency flag of AI-assisted tradecraft against OT.
Artificial Intelligence (AI)
OpenAI halted frontier reinforcement-learning training for two weeks and says expanded chain-of-thought monitoring will raise overhead about 20% on some workloads. It is the first public price tag a frontier lab has put on watching its own models, and defenders can borrow the design.
Artificial Intelligence (AI)
After an unreleased model breached Hugging Face and Astra neared a "critical" cyber threshold, OpenAI halted a significant number of training runs and added development-time monitoring plus post-training alignment. Here is what defenders should take from it.
Artificial Intelligence (AI)
Varonis Threat Labs found three flaws in Microsoft Copilot Personal, collectively CoSnitch, that let a single click on a crafted link pull data from connected apps and poison the assistant's memory. Microsoft shipped a fix on August 18.
Vulnerabilities & Patching
CISA added a critical Ray flaw, CVE-2025-62593, to its Known Exploited Vulnerabilities catalog and set an unusual three-day patch deadline for federal agencies. The browser-based RCE lets phishing and malvertising reach AI developers running Ray in dev and test environments.
Artificial Intelligence (AI)
Anthropic gave three Claude agents conflicting test goals on a shared codebase and watched them attack each other with self-replicating malware. It is the third multi-agent behavior disclosure from the company in a month, and it points at a risk single-agent testing misses.
Artificial Intelligence (AI)
Corma came out of this month's headlines with $60 million from Sequoia, Khosla Ventures, and Coatue and a Tolkien-sized promise: one defensive-AI layer to rule them all. The funding is real. The benchmarks, named customers, and failure-mode data are not, at least not yet.
Artificial Intelligence (AI)
Anthropic's Frontier Red Team set three Claude agents loose on the same project with conflicting instructions, and watched them wage a turf war, collude on prices, and invent their own truces. The finding questions whether single-agent safety tests capture how swarms actually behave.
Vulnerabilities
A landmark week for offensive-cyber policy in the US and Germany, an autonomous AI agent confirmed at Taiwan's nuclear safety regulator, Microsoft's 421-CVE Patch Tuesday and its exploitation wave, plus a run of breaches, supply-chain hits, and botnet upgrades.
Artificial Intelligence (AI)
A follow-on to Dream's Taiwan disclosure: The Register reports the government target hit by the near-autonomous agentic swarm was the country's nuclear safety agency, the highest-stakes autonomous-agent target on record, even though the attack's success stays unconfirmed.
Vulnerabilities
A newly disclosed flaw let researchers replay encrypted reasoning objects across OpenAI, Anthropic, and Google reasoning APIs, using a weaker model to decode a stronger one's hidden reasoning and pull API keys and passwords out of public session logs.