Does Google even want to win at AI?
THE SO WHAT
Debate over whether Google is “losing the AI race” after its DeepMind reorg misses the more important point—incumbents are restructuring to fuse research and product under one roof. For operators, the lesson is governance and integration: your AI efforts will stall if research, infra, and application teams stay siloed.
READ THE SOURCE
MORE FROM THE WIRE
Applied AIMicrosoft kills off unsuccessful AI features while merging its separate Copilot apps
Consolidating consumer and business Copilot into one app — while cutting AI podcasts, Group Chats, Deep Research, and mascot features — is a move toward fewer, higher-usage surfaces. If you’re building on top of Copilot, assume the platform will favor durable workflows over experimental features and design for APIs and entry points that survive these trims.
Applied AIMicrosoft Is Combining Its Copilot AI Into One App That’s Easier to Ignore
One Copilot app instead of a sprawl of clients means less surface area but also fewer forced touchpoints — users can more easily route around it. If you’re betting on assistant distribution, assume you’ll need real pull from specific workflows, not just OS-level placement.
Applied AII looked inside an AI generated movie, and the best parts were all human
Early AI-generated films are landing as scaffolding — structure, filler, and iteration — while audiences still respond most to human-written beats. Studios and creators should treat AI as a volume and exploration tool, not a replacement for the small number of people who can reliably hit emotional tone.
Nvidia is playing many parts in the AI gold rush, a top business guru says
Nvidia acting as picks-and-shovels seller, mine investor, and prospector backer concentrates leverage across the AI value chain. If you’re building infra or platforms, model scenarios where your primary chip supplier is also a co-investor, competitor, and channel — and negotiate accordingly.