Sven Philipsen

Dr. Leonardo Trasande revealed there are over 10,000 chemicals in plastics we know nothing about. “There’s 16,000 chemicals used in plastic materials.” “We know nothing… about 10,000 of them.” “Plastic is a complex and multi-dimensional threat to human health.” “These chemicals hack our hormones.” “We’re talking about impacts on all of us from cradle to grave.”

Anthropic just launched a dedicated enterprise partner network — $100M committed in 2026 for training, support, and co-marketing for businesses adopting Claude. The enterprise AI race isn't just about models. It's about who builds the best adoption infrastructure. #AI

Amazon just gave Prime members free 24/7 access to a personal health AI agent. It can interpret lab results, manage prescription renewals, and book appointments. Democratising healthcare access via AI. That's a big deal. #AI

A new major AI model drops every 72 hours in 2026. Most professionals are still figuring out how to use the one from last month. What does your workflow for staying current actually look like? Curious how others are handling the pace. #AI

85% of teachers and 86% of students used AI in the last school year. The global AI education market will exceed $112 billion by 2034 — up from $7.5B in 2025. Personalised learning at scale isn't a future experiment. It's already in the classroom. #digitalization

AstraZeneca and Tempus AI used contrastive learning to find biomarkers that predict treatment response. Outcome: 15% survival benefit over traditional immuno-oncology trial designs. AI catching what human researchers couldn't. That's not incremental. #AI

Open AI is moving faster than anyone can benchmark. DeepSeek V4: 1T parameters. Open weights. Nemotron 3 Super: enterprise-grade. Open source. One major release every 72 hours globally. The open model moment is just getting started. Here's what that means: 🧵 #AI

AI agents in 2026, in actual numbers: Danfoss: 80% of purchase orders automated, $15M saved/year. Telus: 57,000 employees using agents daily, 38,000 hours saved/month. PepsiCo: 20% throughput increase on first factory deployments. Less hype. More this. #AI #digitalization

Apple and Google quietly confirmed a deal worth ~$1 billion/year. Gemini will power the new Siri — with on-screen awareness, cross-app intelligence, and real context. The world's most-used AI assistant is about to get serious. applemagazine.com/apple-targets-… #AI

Anthropic just made Claude's memory free for everyone — including free users. Even better: you can now import your context and preferences from ChatGPT or Gemini. No starting from scratch. Switching AI assistants just got easier. macrumors.com/2026/03/02/ant… #AI

Nvidia launched Nemotron 3 Super at GTC 2026 — open model, 5× throughput, built for agents. Adobe, Salesforce, and SAP are among the first 17 adopters. Enterprise open AI isn't a future promise. It's here. venturebeat.com/technology/nvi… #AI

GPT-5.4 is "33% less likely to produce false individual claims." That's progress. But "less likely to hallucinate" is still an odd benchmark in 2026 for systems now operating desktops autonomously. The gap between capability and reliability hasn't closed. #AI

OpenAI and Anthropic are now competing in every enterprise deal. OpenAI brings the Pentagon. Anthropic brings Microsoft. If you're making a procurement decision today — does the company's stance on military contracts factor into your choice? #AIGovernance

AI researchers are building simulations of entire human societies — using agents to model real human decision-making in conflict, policy, and markets. The infrastructure for this exists. The ethical frameworks for it don't. nature.com/articles/d4158… #AI

Samsung wants Gemini on 800 million devices by end of 2026. This is what mass-market AI actually looks like — not a frontier model in a data centre, but a mid-range phone in Lagos, Jakarta, or Nairobi. The benchmark wars are the wrong lens. #AI

Gartner: 40% of enterprise apps will have built-in AI agents by end of 2026. That's up from less than 5% today. Most CISOs are not ready for this. Organisations are deploying agents faster than they can secure them. Speed is fine. Governance catching up would be better. #AIAgents

WEF: 170M new jobs from AI by 2030. 92M displaced. Net: +78M. 'Net positive' is real — but only if the transition is managed. The middle part (the 92M) is where policy either delivers or fails. The math works out. The transition doesn't manage itself. #digitalization

Tomorrow is a regulatory inflection point barely making headlines. The FTC must publish its AI policy statement by March 11. Depending on scope, it could pre-empt state AI laws across California, Colorado, and Illinois. That reshapes the compliance landscape for every enterprise AI deployment in North America. #AIPolicy

OpenAI launched GPT-5.4 on March 5 — first model with native computer-use built in. One system that can reason, code, and operate your desktop. That's a meaningful consolidation. openai.com/index/introduc… #AI

Microsoft just announced Copilot Cowork — built with Anthropic's Claude and shipping in M365. OpenAI signed the Pentagon contract. Anthropic refused it. Now Microsoft is betting on Anthropic for enterprise agents. Corporate AI choices are geopolitical now. #AIGovernance