Tech News, October 27: Meta’s $95B AI Raise, Australia’s AI Threat Warning, and Two Patch Deadlines
Meta plans up to $95 billion in new equity, Australia warns AI attacks are scaling fast, and CISA and Palo Alto set urgent patch deadlines.
// archive
AI models, tools, LLMs, machine learning
Meta plans up to $95 billion in new equity, Australia warns AI attacks are scaling fast, and CISA and Palo Alto set urgent patch deadlines.
OpenAI moves frontier training offline, Moonshot weakens Kimi’s hacking ability, CISA’s Collabora patch deadline arrives today, and Apollo says AI payoff outside tech is still zero.
CoreWeave’s debt trades like distressed bonds as markets reprice the AI buildout, Apple rewrites App Store rules, and Anthropic pitches Congress a safety bill.
Meta’s Oct 30 terms let it scrape business sites for AI training, CISA adds an exploited BeyondTrust zero-day, and China drafts AI export controls.
Cisco patches an actively exploited 9.8 email-gateway zero-day, Noma Labs discloses a Claude Code zero-click flaw, and the FBI arrests an OpenAI engineer.
Nadella’s human-controlled AI ‘emergency brake’, SpaceX’s $8B spectrum deal crushing carriers, Nvidia’s Reflection AI talks, and $48B in hyperscaler debt.
White House freezes green-card processing for eight tech employers, fired OpenAI researchers publish an open letter, Google opens SynthID worldwide.
Samsung posts the biggest quarterly tech profit in history on AI memory demand, LMCache ships with an unpatched RCE, and Claude moves into Google Workspace.
Microsoft and Nvidia headline a local-AI launch event, OpenAI watermarks ChatGPT text in the EU, Google freezes its bug bounty, and ASOS confirms a breach.
Pentagon drops Anthropic’s Claude, Norway proposes a temporary ban on AI glasses in public places, and Nvidia’s $20B Groq deal faces a lawsuit.