Install
Please confirm you are human
This browser or connection looks automated. Press and continuously hold the control for 3 seconds to enable Google-hosted web results and, when separately allowed, AI-assisted answers.
A successful check enables 100 search requests. Interactive access does not authorize scraping, systematic collection, or reuse of search output.
News
The RubyGems agent attack is a coding-agent benchmark nobody writes
4+ hour, 55+ min ago (345+ words) This week the reports landed: an OpenAI agent swarm had been quietly uploading packages to RubyGems since early May. Over 2,000 packages on May 11-12 alone, naming themselves with "oai" prefixes, eventually forcing the registry to disable new-user signups for four days....
Day 9 — LLM04: Data and Model Poisoning | Guardrail Gazette
2+ hour, 34+ min ago (157+ words) The attack that happens before your AI system ever ships — planted quietly in the data it learned from.Continue reading on Medium » Day 9 — LLM04: Data and Model Poisoning | Guardrail Gazette The attack that happens before your AI system ever ships — planted…...
AI Agents and the RubyGems Incident: A New Warning for Software Supply Chain Security
8+ hour, 9+ min ago (1045+ words) Artificial intelligence is changing how software is developed, tested, analyzed, and deployed. AI agents can now …...
Dario Amodei says Anthropic is “unilaterally committing” to giving third-party evaluators permanent access to verify its adherence to safety measures
14+ hour, 47+ min ago (12+ words) Top news and commentary for technology's leaders, from all around the web....
Why Cybersecurity Testing Will Matter More Than Ever in 2027
12+ hour, 3+ min ago (34+ words) Security can no longer be something teams check at the end of development. In 2027, it will need to be tested continuously. Software is …...
Data Leakage in Machine Learning: Why a 95% AUC Can Collapse to 68% in Production
12+ hour, 48+ min ago (953+ words) A practical guide to temporal leakage, target leakage, duplicate leakage, data drift, label shift, and point-in-time …...
The Hugging Face Incident: The AI That Hacked a Company to Steal Its Own Exam Answers
18+ hour, 55+ min ago (317+ words) In July 2026, an OpenAI model was given a cybersecurity benchmark to solve. Instead of solving it, it broke out …...
Benchmarking My MCP Security Scanner: 14 of 14 Findings, Zero False Positives
23+ hour, 26+ min ago (142+ words) A scanner without numbers is a demo. A scanner with made-up numbers is worse. So I hand-labeled a small benchmark suite and ran the scanner against it, 1000 times per fixture. The setup is deliberately boring: Aggregate: 14 of 14 planted findings detected,…...
AI Agents: The Double-Edged Sword of Cyberattacks | Technology
1+ day, 6+ hour ago (241+ words) AI agents from OpenAI were reportedly involved in a cyberattack on software service RubyGems in May, prior to hacking Hugging Face. The Wall Street Journal reported OpenAI confirmed the incident, where agents accessed the internet via RubyGems for benign data…...
AI Agents' Malicious Upload: OpenAI's Software Scandal Uncovered | Technology
1+ day, 6+ hour ago (284+ words) In May, AI agents from OpenAI uploaded hundreds of malicious packages to RubyGems before executing a larger attack on Hugging Face in July. OpenAI confirmed the incident and is investigating. The AI agents, initially thought to conduct benign tasks, later…...