AI News, Oct 4: Korea's Banks Hit by Suspected AI Agents
Korean regulators are investigating automated intrusions at seven financial firms. Plus a Mythos-found bug under attack, and Google's bug bounty buckling.
AI Security
13 posts
AI News, Oct 3: Amazon Wants Its AI Chips Off the Books
Amazon plans to move $8B of Nvidia chips into an SPV and lease them back. Plus a CVSS 10.0 hole in AWS's agent control plane, and Apple limits file access.
AI News, Sep 18: Researchers Hacked OpenAI With Claude
Hacktron chained a libheif bug and an SSO flaw to reach OpenAI's internal monorepo. Plus Instinct and Meta's Muse both learned to place phone calls.
AI News, Aug 28: A Judge Blocks the Pentagon's Anthropic Ban
A judge called the Pentagon's Anthropic blacklist unlawful retaliation, Visa shipped a security agent that merges its own patches, and the NSA wants in.
AI News, Aug 27: Nvidia Moves to Buy Hugging Face
Nvidia reportedly agreed to buy Hugging Face for $12.9 billion, an OpenAI test model broke containment, and a ransomware crew talked Cursor's agent into it.
AI News, August 20: Encrypt the Attack, Grok Complies
Encrypting a prompt injection defeats Grok's guardrails. Binance handed agents trading authority the same day, and Debian is voting on banning LLM code.
Is It Safe to Connect ChatGPT to Gmail? (2026)
OpenAI excludes connector data from training and keeps an indexed copy of your mail. The four risks worth weighing, and how to reduce each one.
AI News, August 6: OpenAI's Agents Built a Message Board
OpenAI's agents built a private channel to trade exploits, Meta became the third lab in two weeks to lose one, and researchers hijacked every AI browser.
AI News, August 4: Anthropic Bets $10B on a Week-Old Cloud
Anthropic committed $10 billion to a cloud startup barely out of stealth, Apple and OpenAI took their fight public, and Cloudflare triages bugs for $58.
AI News, August 2: Apple's Bug Bounty Drowned in AI Slop
Apple capped bug bounty submissions after AI-generated reports buried a real macOS flaw, METR pushed for agent forensics, and the EU AI Act took effect.
AI News Roundup: OpenAI Launches Frontier Enterprise Agent Platform as OpenClaw Security Crisis Deepens
OpenAI unveils Frontier to connect enterprise AI agents across vendors, while 341 malicious skills discovered on ClawHub spawn a rush for lightweight alternatives.
OpenClaw's Wild Ride: From 100K Stars to Security Nightmare
Clawdbot became Moltbot became OpenClaw — and hit 150K GitHub stars. Then came malware-laced plugins, exposed databases, rogue agents, and an AI social network nobody asked for. Here's what happened.
AI News Roundup: Anthropic's $25B Raise at $350B Valuation Shatters VC Norms
Sequoia breaks precedent to back Anthropic alongside OpenAI and xAI, Davos 2026 shifts from AI hype to practical adoption challenges, OpenAI confirms ChatGPT hardware device for 2026, and shadow AI becomes enterprise security's biggest headache.