Google Tried to Cut School Search Access Over AI Restrictions

The AI industry is having a very bad Tuesday, and not just because of election results.

Google Just Tried to Bully Schools Into Using Its AI

Google reportedly planned to cut off access to its search engine for schools that blocked or restricted its AI tools — meaning a district that told students "no AI on assignments" could have lost the ability to Google anything at all. NYC Mayor Zohran Mamdani and former FTC chair Lina Khan called it what it looks like: a monopoly using its grip on one essential service to force adoption of another. Every school district that's built its curriculum around Google's free tools just got a reminder of what "free" actually costs. The backlash was loud enough to generate political heat within 24 hours — Google badly underestimated how much harder it is to push AI on parents and school boards than on enterprise procurement teams.

https://techcrunch.com/2026/11/04/google-faces-backlash-for-plan-that-would-turn-off-search-for-schools-that-restrict-ai/

Voters Just Put a Wall Around the AI Industry's Power Supply

Data centers — the warehouse-sized facilities packed with servers that actually run AI — lost big in yesterday's midterms. Candidates who campaigned on moratoriums and new restrictions on data center construction won across multiple states and localities. This matters because the electricity and land those facilities require have been expanding at a pace that local communities are increasingly unwilling to accept. If you work in any industry that's been promised AI-powered efficiency gains by 2027 or 2028, those timelines just got less certain. AI companies have been treating "can we get permits for this" as a footnote. It's not a footnote anymore.

https://www.theverge.com/policy/1026365/data-centers-midterms-results

AI Agents Booked Uber Rides Four Times Last Quarter. Total.

There's been enormous hype around "AI agents" — software that can take actions on your behalf, like booking a car or ordering dinner, without you touching an app. Uber disclosed this morning that in Q3 2026, AI agents arranged exactly four rides through its platform. Four. The food delivery numbers were higher but still negligible. This is worth sitting with for anyone whose job involves evaluating AI vendors or setting expectations with leadership: the gap between what agents can demo and what they actually do at scale, in real commercial environments, is still enormous. Your vendor's roadmap probably has "agents handling X by Q2" somewhere in it. Uber just told you what Q3 2026 actually looked like.

https://www.theverge.com/transportation/1026306/uber-ai-agents-q3-earnings-disintermediation

The UK's £400 Billion AI Promise Was Basically a Google Press Release

The British government cited a figure of £400 billion in economic gains from AI when making the case for its national AI strategy. The AI minister now admits the real estimate is less than a third of that — and investigators found the original number came primarily from Google's own projections, which the government apparently adopted without serious scrutiny. AI economic forecasts frequently originate with companies that benefit from large forecasts, get cited by politicians who didn't bother to check the sourcing, and then quietly get revised down after the policy is set. If your organization is making budget or hiring decisions based on published AI productivity numbers, ask your vendor to show you the original study — not a summary — and check the acknowledgments section for who paid for it.

https://www.wired.com/story/uk-officials-said-ai-would-add-400-billion-to-the-economy-that-figure-mainly-came-from-google/

Sam Altman Spent Election Night Arguing About RAM

OpenAI's CEO changed his display name on X to "Ram Altman" last night after getting into a public back-and-forth about whether consumers should buy more RAM — the short-term memory inside a computer that helps it run multiple things at once — for their personal computers. The original comments apparently implied people were wasting money on RAM purchases. Altman pushed back, saying he wants people to have as much memory in their computers as they want. This is minor, but it's the kind of thing that matters for reading the room: the CEO of the most prominent AI company spent election night, the same night his industry took a serious political hit on data centers, in a Twitter argument about PC hardware.

https://www.theverge.com/ai-artificial-intelligence/1026292/sam-altman-ram-shortage-pcs-comments

The industry that promised to move fast is now running into voters, school boards, and actual usage numbers — and it turns out all three move at a very different speed than a product announcement.

← More daily briefs