Strategy & Market Jensen Huang Resigns: Nvidia Faces a Leadership Test Jensen Huang has resigned from Nvidia. Colette Kress is interim CEO; the company has not stated a reason, and espionage claims remain unproven.
Strategy & Market Cadence Maestro Makes Chip Design Models a Portfolio You Post-Train Cadence opened its Maestro portfolio of domain-specialized models for chip and system design, post-trainable on a customer's own silicon data and runnable on premises, with first access in t
Strategy & Market Google TPUs Land on Oracle Cloud and Accelerator Choice Becomes a Cloud Feature Oracle will offer Google TPUs through OCI in calendar year 2027, the first hyperscale cloud outside Google to do it, with Oracle supplying the power, cooling and datacenter networking.
Strategy & Market How a Website Becomes Citable by Search Engines and AI Assistants A site gets cited when a crawler can fetch the page, read the text in the HTML it receives, attribute it to a named organisation and confirm the facts in a structured data block. This is the
Strategy & Market Anthropic Turns AI Adoption Into a Training Programme for Customer Engineers Anthropic launched Claude Frontier Academy on 2 October 2026 with a $100 million commitment and a target of 10,000 Frontier Deployed Engineers by the end of 2027. The programme pairs a four-
Sovereignty & Compliance China Proposes Export Controls for Large AI Models China's commerce and science ministries published a draft revision on 16 October 2026 that adds 40 technologies to the export catalogue, 23 of them AI. Models above a two-trillion-par
Strategy & Market The World Bank Wants Small AI Built for Weak Connections The World Bank's chief knowledge officer argues that narrow models running on ordinary phones, in local languages and without reliable connectivity, can shorten development timelines. The Se
Sovereignty & Compliance CERN and Mistral Plan Flare, a Foundation Model for Collider Physics CERN and Mistral signed a statement of intent on 15 October 2026 to build Flare, a single pretrained model for collider physics trained on billions of simulated LHC collisions, with no budge
Strategy & Market Tesla Flying Car: How the Roadster Rocket Uses Compressed Air The Roadster Rocket is Tesla’s one-seat, compressed-air-thruster version. Learn how it works, what the reveal showed and which flight details remain open.
Sovereignty & Compliance A Lobby-Funded Study Prices Europe's Cloud Rules at €115 Billion ECIPE published a study on 13 October 2026 that puts ten years of adjustment costs from Europe's Cloud and AI Development Act at €115 billion in its central scenario, with a €233 billion ver
Sovereignty & Compliance Four States Put TP-Link's Router Security Claims on the Record Four state attorneys general sued TP-Link over router security claims on 6 October, the vendor published a defence the same day, and a 21-state coalition took the same argument to the FCC th
Sovereignty & Compliance The 2026 State of Open Source in Europe Report Puts Sovereignty in Numbers The fifth State of Open Source in Europe report finds 94% of European organisations treat digital sovereignty as strategically important, and it prices what governance and private forks cost
Sovereignty & Compliance NVIDIA Puts $1 Billion and Seven Supercomputers Behind the Genesis Mission A $1 billion commitment, seven new systems and one AI-first design choice: NVIDIA's Genesis Mission work shows where scientific computing trades precision for throughput.
Strategy & Market SAP Starts Its Joule Work Rollout With Its Own Numbers Attached SAP opened the customer rollout of Joule Work on 6 October 2026 with 110,000 of its own employees already working in it, and published the productivity numbers it measured internally.
Sovereignty & Compliance Europe Starts a 24-Hour Clock for Exploited Vulnerabilities Article 14 of the Cyber Resilience Act has applied since 11 September 2026: manufacturers report actively exploited vulnerabilities and severe incidents within 24 hours, then 72 hours, then
Strategy & Market Deutsche Telekom's €2.5 Billion AI Target Is a Measurement Design Deutsche Telekom says AI and automation will remove about 2.5 billion euros of indirect cost by 2030 against a 2023 baseline. The euro figure is a forecast; the unit metrics published alongs
Sovereignty & Compliance Europe's Sovereignty Package Draws a Line Most Clouds Will Not Cross The Commission's Technological Sovereignty Package — Chips Act 2.0 and the Cloud and AI Development Act — grades cloud services on a four-level sovereignty scale. Defence-pushback shows the
Strategy & Market The AI Labs' Public Pacing Calls Are Now Antitrust Evidence Four subscribers filed a class action against Anthropic, OpenAI, SpaceXAI and Google over the September 12 pacing coordination. Netics reads the complaint: what it gets right, what it conced
Strategy & Market OpenAI Ads Manager Campaign Architecture: Budgets, Bids, and Measurement Contracts OpenAI Ads Manager exposes a campaign hierarchy. Netics explains why its IDs, budgets, bids, creative, and tracking rules form a control contract.
Strategy & Market OpenAI Ads Manager Moves Campaign Operations Into an Operating Layer OpenAI is testing Sponsored Agents and adding natural-language campaign tools, but the useful shift is operational: review, labeling, privacy, and approval still define the boundary.
Strategy & Market ChatGPT Ads measurement needs an explicit boundary OpenAI is adding self-serve buying and conversion tools to ChatGPT Ads. Netics argues that measurement is useful only when conversion and privacy boundaries are explicit.
Strategy & Market Apple’s Hybrid AI Strategy Puts the Control Plane at the Center Apple is not abandoning apps or services. Its more durable AI strategy is a hybrid control plane that routes work across the device, Private Cloud Compute, and external models.
Strategy & Market NVIDIA Buys Hugging Face: Can Open AI Stay Open Under Chipmaker Ownership? NVIDIA says Hugging Face will stay open. Netics examines whether provider neutrality and portability can survive chipmaker ownership.
Sovereignty & Compliance Europe's AI Sovereignty Problem Isn't Model Access — It's the Skills to Run What You Deploy Europe's 2026 tech talent data shows the real AI bottleneck: 48% cybersecurity understaffing and 32% Ray deployment, not model licensing.
Strategy & Market Production AI Starts Beyond the Model A working AI model is not production-ready until identity, spend, capacity, observability, permissions, and rollback are controlled.