🌍 US Locks Out Allies • 🧬 Google's Ultimate Brain Drain • 🎬 Amazon Censors Altman
Why Amazon just walked away from a near-complete $40M biopic to protect its multi-billion-dollar OpenAI alliance.
🎵 Podcast
Don’t feel like reading? Listen to it instead.
P.s. I know this looks broken in email…
📰 Latest News
This week’s image aesthetic (Flux 2 Pro): 1970s Corporate Surrealism
The G7 Tech Bloc: US Tightens Its Grip on Frontier AI
The fallout from the US government’s sudden export ban on Anthropic’s Fable 5 and Mythos 5 models has escalated into a major diplomatic headache. The unilateral directive, which severed access for all non-US citizens globally over fears that the models’ cyber-offensive safeguards could be bypassed, dominated discussions at the recent G7 summit.
British Prime Minister Keir Starmer reportedly lobbied the White House for a country-specific exemption for UK firms, only to be told there was “zero chance” of an arrangement. Meanwhile, Anthropic’s international head, Chris Ciauri, is engaged in urgent damage control in Seoul, publicly predicting that access will be restored to South Korean enterprises within days.
Seizing on the chaos, the CEOs of Anthropic and Google DeepMind used a private G7 session to pitch a radical solution. They proposed a formal, US-led coalition of democratic nations to strictly govern access to advanced models, coordinate cyber-defence, and completely choke off the supply of microchips to China.
Why it Matters
This is the moment commercial artificial intelligence formally became an instrument of state-directed geopolitics. The abrupt blacking-out of Anthropic’s models across 15 countries proved to allied nations that relying on American tech companies leaves their domestic infrastructure highly vulnerable to Washington’s regulatory whims.
The fact that the UK was flatly denied an exemption underscores a brutal reality: in the eyes of US national security officials, there are no “trusted allies” when it comes to restricting digital weapons. By proposing a formal democratic tech bloc, tech executives are attempting to front-run this protectionist shift, effectively carving the global technology sector into rigid, cold-war-style alliances.
🔗 More from The Telegraph - British carve out
🔗 More from CNBC - G7 coalition
The Great Google Brain Drain: DeepMind Loses Its Crown Jewels
Google’s research division is facing an unprecedented talent crisis following the high-profile defections of two of its most celebrated pioneers. John Jumper, the 2024 Nobel Prize winner who famously led the development of the revolutionary AlphaFold protein-prediction system, has officially left Google DeepMind to join rival lab Anthropic.
His exit closely follows that of Noam Shazeer, a principal co-author of the foundational 2017 “Attention Is All You Need” transformer paper and former co-lead of Google’s Gemini models, who has jumped ship to OpenAI.
Why it Matters
Losing the literal architects of your core technology to your primary commercial rivals is a catastrophic blow for Google. Shazeer and Jumper represent the absolute pinnacle of AI engineering and computational biology.
Their departures prove that retaining elite talent in this hyper-competitive landscape requires far more than Google’s deep pockets. As Anthropic aggressively weaponises Jumper’s expertise to expand into life sciences, Google is watching its historical dominance in scientific AI completely evaporate.
The Hollywood Blacklist: Amazon Drops the Altman Biopic
Amazon MGM Studios has abruptly halted the distribution of Artificial, a highly anticipated biographical feature film detailing the dramatic 2023 ousting and reinstatement of OpenAI CEO Sam Altman. Directed by Luca Guadagnino and starring Andrew Garfield, the project was nearly complete when Amazon walked away, leaving a 40 million USD investment in limbo while it searches for a new distributor. The sudden cancellation occurred just months after Amazon finalised a massive, historic strategic alliance with OpenAI.
Why it Matters
The sudden death of Artificial which insiders report offered a deeply unflattering depiction of Altman exposes the toxic overlap between tech monopolies and modern media. Amazon’s multi-billion-dollar corporate alliances now directly dictate what narratives are allowed to reach public theatres. The fact that filmmakers were left completely blindsided suggests that when the business interests of cloud infrastructure conflict with creative expression, corporate tech giants will quietly censor high-profile media to protect their software partners.
Pay-Per-Task: Microsoft Legalises the Digital Employee
Microsoft has finally launched Copilot Cowork for all global Microsoft 365 enterprise subscribers. Moving past standard chat assistants that merely summarize emails or write text, Cowork is a fully autonomous digital teammate engineered to execute complex, multi-step workflows across applications like Outlook and Teams from a single natural language prompt.
Crucially, the tool abandons flat-rate subscription models in favour of a usage-based consumption framework, forcing companies to buy “Copilot Credits” to pay for individual automated tasks.
Why it Matters
This marks the official corporate transition from conversational AI to autonomous agentic workforce deployment. By allowing software to independently execute entire cross-application workflows, Microsoft is shifting the definition of automation from an administrative aid to an active digital worker.
The introduction of “Copilot Credits” is a massive commercial pivot. It forces enterprise procurement departments to budget for AI software exactly like human independent contractors, tying the financial return on investment directly to the volume of completed operational tasks rather than per-seat licensing.
Last week’s newsletter:
