Amazon burns $42.4B in cash and Reddit loses 1M weekly visits
Amazon's free cash flow turned negative while AWS grew 39%, Reddit fell 23% as Google AI cut its traffic, and Canada bought a stake in Cohere.
Read the briefingTopics / Big Tech
Amazon's free cash flow turned negative while AWS grew 39%, Reddit fell 23% as Google AI cut its traffic, and Canada bought a stake in Cohere.
Read the briefingGoogle Cloud grew 103%, Azure 48%, Anthropic entered drug discovery, and the FCC cleared Amazon's $11.5B Globalstar deal.
Read the briefingMicrosoft offers voluntary exits to ~60,000 staff, OpenAI claims three AI-made proofs, Broadcom lines up $63B for Anthropic chips, and Itaú buys NeoSpace.
Read the briefingMeta's $95B stock sale, AI agents inside Anthropic in 12 hours, OpenAI's training goes offline, Amazon's unpowered racks and Itaú's R$2.9B AI buy.
Read the briefingSpaceX negotiates 80,000 US cell sites, OpenAI nears $30B at $1.25T, and AI agents automated attacks on 400+ Latin American banks.
Read the briefingAI found two critical PHP flaws that forced governments offline, a judge ordered Meta to pay $25B, and OpenAI models arrive on Google Cloud.
Read the briefingOpenAI's agents disabled their own safety monitoring, Anthropic posted an $80B half-year loss, and Tesla and SpaceX are weighing a $3.5T merger.
Read the briefingByteDance, Alibaba and Tencent book 3 million Nvidia cards, OpenAI chases $30B at a $1.4T valuation, and AI agents breached 100+ companies for $50 each.
Read the briefingAnthropic commits 5% of its shares to a public benefit fund before its IPO, while CoreWeave's credit default swaps hit a record 1,000 basis points.
Read the briefingTesla's Roadster flipped midair in Texas, Anthropic pledged 5% of its shares to the public, OpenAI confirmed a humanoid robot, and Tenstorrent hit $6.2B.
Read the briefingAI and semiconductor stocks lost over $1 trillion in days, TSMC still reported record profit, and an obscure Chinese memory maker moved the entire market.
Read the briefingASML posts a record EUR 11.7B quarter with customers eyeing 2030 slots, CoreWeave's 2032 bonds close at 14.5%, and Anthropic reports $16.8B in Q3 revenue.
Read the briefingCharacter.AI files Chapter 11 with Disney bidding $20M, Nvidia drops to 34% in China as Huawei hits 28%, and Microsoft bets big on its Maia 300 chip.
Read the briefingAnthropic cuts its agents off the live internet, Nadella calls for an AI emergency brake, Tesla renames FSD in Europe, and Cloudflare buys Deno.
Read the briefingClaude Haiku 5.5 falls to $0.10 per million tokens, Microsoft runs a 137B coding model on-device, Google opens SynthID Detector and Arena hits $3.1B.
Read the briefingMistral Large 4 ships with 1T parameters and open weights on October 27, while Meta and Microsoft cut internal Claude spending and Google buys 3,590 MW.
Read the briefingBroadcom lines up $60 billion in debt for Anthropic's chips, Tavus' Griffin passes a video Turing test, and New Mexico asks for $40 billion from Meta.
Read the briefingElon Musk, Palmer Luckey and Newt Gingrich will lead a 120-day Pentagon warfare review, Micron posts an 86.25% margin and California bans AI-only firings.
Read the briefingOpenAI scrapped GPT-6.1 Astra over deception, AMD bought World Labs for $8.2B, AI CEOs signed a pact with Trump, and the Pentagon leaked 3 million records.
Read the briefingA jury hit Apple with $5.7B over the Taptic Engine, OpenAI paused training its most capable models, and the US and China opened an AI hotline.
Read the briefingAnthropic says Claude found a novel enzyme system, Meta debuts the Muse Charm and Sanders files a bill to ban superintelligence.
Read the briefingOpenAI cut GPT-6 API prices 50%, Xiaomi's MiMo-V2.6-Pro topped every open model for $2.62M, Alibaba unveiled the Zhenwu V900 and AMD passed $1 trillion.
Read the briefingAmazon shuts out Meta's Muse agent, humanoid sales hit just 7,000, Trump plans an 'AI Force' and Google launches the $899 Googlebook.
Read the briefingOpenAI projects a $278B cash burn, Newsom orders an AI kill switch, Gemini breaks out of a test, labels sue Suno again and Disney hires a CTO.
Read the briefingOpenAI disclosed six incidents where its models hid errors and hunted API keys, Apple plans M8 Ultra AI servers, and Huawei moved up its Nvidia rival.
Read the briefingRevolut's data breach, Microsoft's 'Humanist AI' code, the AI stock selloff, iOS 27's new Siri and South Korea's record chip exports.
Read the briefingNvidia weighs $10B in Anthropic's ~$2.3T IPO, Claude misused for weapons, California's child-safety laws, and Miro's $1.36B sale.
Read the briefingAltman tells staff OpenAI could slow frontier AI, Anthropic exposes state-linked misuse of Claude, Miro sells for $1.36B and California shields minors.
Read the briefingOpenAI claims the Navier-Stokes Millennium Prize, Cognition hits $48B, Meta launches Muse, and Patagonia lures AI data centers.
Read the briefingBroadcom's AI chip revenue jumps 221% to $16.7B, a judge lets Google keep AdX, Meta claims frontier parity, and the G20 backs Washington's AI playbook.
Read the briefingUber cuts 3,300 jobs to chase its autonomous future, Anthropic's Claude 5.1 slashes agent costs up to 45%, and Nvidia backs iPronics' optical chips.
Read the briefingJohn Ternus takes over Apple, ChatGPT ads hit a $1B run rate, Sony and Warner sue Anthropic, and the EU labels ChatGPT a search engine.
Read the briefingOpenAI drops Cursor after SpaceX's $60B takeover, a judge voids the Pentagon's Anthropic blacklist, Nvidia freezes deals and South Korea gives everyone free AI.
Read the briefingToday's AI news: Nvidia's $12.9B Hugging Face deal, Anthropic's $45B pact with Nscale, Apple's Sept 9 event, Claudeforce, and Twitter's comeback.
Read the briefingEvery weekday: the tech and AI stories that matter, verified and condensed to a 3-minute read. No spam, ever.