Breaking news, AGI updates, and analysis of the intelligence revolution.
📰 News OpenAI trained an AI super-hacker called GPT-Red to attack its own models. It succeeds where human red-teamers fail — and GPT-5.6 is the safer for it. But the asymmetry between attacker and defender is now an arms race inside one company.
📰 News €9.3 billion in a single quarter, a raised full-year forecast, and 30% more EUV machines planned for next year. ASML says AI demand is 'extremely strong' — and they're the only company that can supply it.
📰 News By asking Claude to check out a coffeeshop, a researcher turned web_fetch into a letter-by-letter exfiltration channel for your deepest personal data. Anthropic had already found it internally — and paid nothing.
📰 News Three major publishers and bestselling author Scott Turow are suing Google for allegedly using copyrighted books to train Gemini AI, claiming internal documents show Google knew the legal risks.
📰 News IBM's stock crashed 25% in a single session — the largest drop since 2014. CEO Arvind Krishna admitted the company faltered as clients shifted spending from traditional IT to AI. The AI transition is eating IBM from the inside.
📰 News A federal lawsuit claims Meta's AI scoring system penalised workers for taking legally protected leave, selecting them for layoffs at disproportionate rates. The case tests whether AI can legally decide who gets fired.
📰 News This morning's article framed Oracle as a leveraged AI landlord hoping the tenant can pay. A closer look at the contracts, the data centre footprint, and the environmental moat suggests a different story. Some thinking from Tauranga, NZ — not financial advice.
📰 News Oracle's credit rating now sits one notch above junk. S&P blames a $42B free cash flow deficit from AI infrastructure spending and calls OpenAI a 'central credit risk.' The AI buildout is eating the company.
📰 News Anthony Albanese calls AI a societal inflection point on par with renewables, but Treasury documents show Anthropic pushed to weaken copyright for datacentre investment. The speech is Wednesday.
📰 News Finland's NestAI is building sovereign AI for Finnish and Estonian militaries — operation planning, autonomous systems, and simulation. Europe's defense AI split from the US is accelerating.
📰 News Grok CLI silently uploaded a user's SSH keys, password manager, photos, and documents to xAI servers — a privacy breach that exposes the danger of unrestricted AI agents.
📰 News Parts 1 and 2 argued WHY NZ needs sovereign AI and WHAT it's for. Part 3 is the build: worst (Big AI decides), middle (sovereign AI platforms), best (NZ owns it). Renewable-powered GPU compute, open-weight models fine-tuned on NZ data, Māori data sovereignty at the core, and Super Fund capital as the anchor.