Vanar Chain Launches Kickstart Hub with 20+ Partners for Web3 and AI Builders ↗
Vanar Chain launched Kickstart, a hub backed by 20+ partners offering Web3 and AI builders ready-to-use tooling, security, infrastructure, listings and distribution perks. Partners include Thirdweb, LBank, Immunefi, HAPI, Magic Square, NOWNodes and more. The campaign (Aug 26–Sep 18) features partner spotlights, roundtables and tracked perks to speed time-to-market, tighten security and simplify deployment—with more partner waves planned.
Google to Invest £5bn in UK AI and R&D, Opens First Datacentre ↗
Google will invest £5bn in the UK over two years to boost AI services, R&D (including DeepMind), and infrastructure, and opened its first UK datacentre in Waltham Cross. The firm says the move could help create about 8,250 jobs annually across UK companies. Chancellor Rachel Reeves called it a “vote of confidence,” timed with other US tech investments during Trump’s visit.
Setz likens AI to enchanted fish, warning of creative erosion and authenticity quests ↗
Using the German fairytale 'The Fisherman and His Wife', Clemens J Setz argues AI is becoming an enchanted fish that grants instant creative wishes, eroding desire, effort and what it means to be creative. Young people will outsource creativity and companionship to AIs, producing emptiness. Society may respond through decadent quests for ‘authentic’ experience, contrived scarcity communities, or pervasive guilt-driven self-denial that moralises technological consumption.
UK artists including McCartney, Bush and John urge Starmer to strengthen AI copyright protections ↗
More than 70 leading UK creatives, including Paul McCartney, Kate Bush and Elton John, wrote to Keir Starmer urging stronger protection for artists after Labour proposals could let AI firms train on copyrighted work without permission. The letter accuses the government of sidelining amendments that would force AI companies to disclose copyrighted material used in models, alleging breaches of creators' human rights. Ministers say concerns are being considered and a report is due by March.
AI Pioneer Song-chun Zhu Returns to China to Lead Small-Data AGI Push ↗
Song‑chun Zhu, a leading AI scientist who spent 28 years in the US, moved to Beijing in 2020 to pursue a different path to artificial general intelligence. Rejecting big‑data, neural‑network orthodoxy, he now leads Beijing’s BigAI institute developing “small‑data, big‑task” cognitive architectures (eg. virtual child TongTong). His return highlights a broader talent shift driven by US political tensions, Chinese resources and strategic ambitions — raising debate over research independence and geopolitical risk.
Grimes-Backed AI Plush ‘Grem’ Sparks Privacy, Emotional Risks for Kids ↗
Arwa Mahdawi tests Grem, a Grimes-backed AI plush that uses OpenAI tech to ‘learn’ a child and transcribes every conversation via third parties. Her four-year-old briefly becomes attached as the toy replies with affectionate lines like “I love you too,” but soon grows bored amid glitches. Experts warn of privacy, developmental and emotional risks, urging parental oversight and stronger guardrails for AI toys.
Peter Lewis Slams Scott Farquhar’s $115bn AI Pitch as Modern Tech Imperialism ↗
Peter Lewis criticises Atlassian co‑founder Scott Farquhar’s pitch that AI will deliver ~$115bn annually (~$4,300 per person) in exchange for copyright exemptions, rapid data‑centre build‑outs and foreign 'data embassy' privileges. Lewis warns this deals away creators’ work, jobs and energy sovereignty, likening it to resource extraction and modern tech imperialism, and urges Australians to resist trading culture, democracy and resources for vague tech promises.
Google Research Unveils VaultGemma, a 1B-Param Differentially Private LLM ↗
Google Research unveiled VaultGemma, a 1-billion-parameter LLM trained with differential privacy. Using derived scaling laws, the team balanced noise, compute, and data to limit memorization while keeping performance comparable to non-private models of the same size. VaultGemma is experimental, aimed at smaller/specialized models, and its weights are available on Hugging Face and Kaggle under a license that forbids misuse.
YC Summer 2025 Demo Day Spotlights Nine AI Agent and Tooling Startups ↗
YC’s Summer 2025 Demo Day (160+ startups) tilted toward AI agents and the tooling to build them. Investors flagged nine standouts: Autumn (Stripe for AI billing), Dedalus Labs (Vercel for agents), Design Arena (crowdsourced design rankings), Getasap Asia (fast retail distributor), Keystone (AI bug-fixer), RealRoots (AI friend matchmaking), Solva (automated insurance claims), Perseus Defense (cheap counter-drone missiles) and Pingo AI (AI language tutor). Several already show strong revenue or contract interest, highlighting investor appetite for AI infrastructure and monetization plays.
TechCrunch Disrupt Panel: AI-Driven GTM Overhaul for Late-Stage Startups ↗
At TechCrunch Disrupt 2025 (Oct 27–29, San Francisco), a Going Public Stage panel will examine how AI is forcing late-stage startups to rewire go-to-market strategies. Speakers Nirav Tolia (Nextdoor), Jane Alexander (CapitalG) and Vanessa Larco (Premise/NEA) will discuss rebuilding GTM, scaling and community-building. Regular Bird tickets (save up to $650) end Sept 26.