tracking datacenters and the politics & regulation shaping AI @semianalysis_
Joined January 2016
- Tweets38
- Following367
- Followers1.4K
- Likes471
Glad our analysis was compelling enough to revise your forcast📈
U guys do great work. Fun to chop it up w @dylan522p at our AGM this week. After listening to the @elonmusk call, your convo, & my Jensen pod last week - our forecast likely trending higher. 🧐📈
Maya Barkin retweeted
According to security officials, pro-Russian and far-right social media accounts linked to previous disinformation campaigns have spread violent videos in an attempt to sow more chaos ft.trib.al/t4nxjpO
ALT A protester wrapped in a French national flag stands in front of Lyon's tourism office set on fire during a demonstration in Lyon, southeastern France. Text on image reads: EU probes pro-Kremlin disinformation around French riots (Credit: Jeff Pachoud/AFP/Getty Images)
Maya Barkin retweeted
Is 50 GW in a single year too aggressive?
"Dear Brad called us out publicly and cut our forecast in half."
So Jeremie (@JeremieEO) and Reyk (@robotknower) came on with Jordan (@JordanNanos) to show their work. They go site by site from satellite and explain why haircutting the number double counts the delays already baked in. Plus moratoriums, 9-month modular builds, and why 2028 is the year to watch.
Maya Barkin retweeted
🇩🇪 Nvidia's Jensen Huang CORNERED German Chancellor Merz on the old EU regulation playbook
"I've also heard you say that you want to find that balance between innovation and regulation.
There's nothing to regulate if the industry doesn't thrive."
Europe keeps reaching for the rulebook before the companies even get off the ground.
Writer: Lucas
Datacenters have become a major US campaign issue and moratoriums are spreading fast, but the political noise far outweighs the megawatts at stake. Just ~2.3 GW of ~21.2 GW in affected areas (ex-Texas) is at real risk of delay, as @TheEconomist highlighted this week citing my research at @SemiAnalysis_. November could change that if anti-datacenter candidates win and the rules tighten, which is why we track these projects one by one.
Maya Barkin retweeted
Brazil sees record fall in Amazon deforestation as key election looms bbc.in/4B8deRb
Maya Barkin retweeted
Moratoriums aren't killing the US datacenter buildout.
Our analyst @mayabarkin finds only ~2.3 GW of ~21 GW in moratorium areas (excluding the TCEQ pause in Texas) is likely to be delayed, as cited in The Economist.
For more, read her report in the SemiAnalysis newsletter👇️
newsletter.semianalysis.com/…
xitter.cf/ArchieHall/status/2111…
A few reflections from travelling (w/ @ed_shackle, @sebwride++) through data-centre country in Michigan/Ohio in my latest for @TheEconomist.
1/ The issue really is moving votes to the Democrats. That's now visible in our polling, too (h/t @OwenWntr), but also in the scale of on-the-ground discontent. I'm quite a bit more bearish on the prospects for big new data-centre projects in blue states after the trip.
2/ I really didn't hear that much direct anti-AI sentiment. The mood much more was a cocktail of worries about local harms (water, electricity bills, house prices), plus a total distrust of businesses and politicians. Against that backdrop, assurances about the water usage of closed-loop cooling systems really don't cut through. And, though developers will doubtless get cannier at this, cash payoffs to communities can easily look like subterfuge and do little to help.
3/ Still, moratoriums are at this stage much more bark than bite (h/t @mayabarkin for the data there). Only about 2GW seems actually delayed. The bigger risk is deterring newer projects, but I'd still guess that affects the exact locations more than whether the build-out happens at all.
4/ The question I'm left with is what comes after data centres as the big AI political issue into next year. A few obvious candidates are: a big cyberattack, the sheer scale of AI-related PAC spending in the 2028 cycle, the early innings of job displacement.
5/ What the data-centre issue should tell us, I think, is that voters' hostility to new technologies that change their lives in unpleasant ways is vast—especially when those changes are less positive-sum than data centres. So, however mad you expect AI as a technology to make the next few years, scale that up accordingly to account for a much more scrambled politics.
Do take a look at my piece on this here: economist.com/united-states/…
And at our Checks and Balances podcast, where I contributed a short diary featuring lots of conversations with voters economist.com/podcasts/2026/…
Maya Barkin retweeted
Mais de 100 organizações acusam big techs de interferir nas eleições brasileiras
www1.folha.uol.com.br/coluna…
Maya Barkin retweeted
US election interference:
Brazil’s Public Ministry investigates Meta over apparent manipulation of presidential race.
brasilwire.com/us-election-i…
Happy to see my research on the impact of datacenter moratoriums making it into @TheEconomist.
Despite the headlines, the rapid spread of moratoriums across the US has had little effect on actual buildout. Roughly 21.2 GW of capacity sits in areas covered by local or state moratoriums (excluding Texas), yet only ~2.3 GW of that is likely to slip. The political momentum behind these fights is real, but for now the capacity at stake is small.
Full breakdown at @SemiAnalysis_.
A few reflections from travelling (w/ @ed_shackle, @sebwride++) through data-centre country in Michigan/Ohio in my latest for @TheEconomist.
1/ The issue really is moving votes to the Democrats. That's now visible in our polling, too (h/t @OwenWntr), but also in the scale of on-the-ground discontent. I'm quite a bit more bearish on the prospects for big new data-centre projects in blue states after the trip.
2/ I really didn't hear that much direct anti-AI sentiment. The mood much more was a cocktail of worries about local harms (water, electricity bills, house prices), plus a total distrust of businesses and politicians. Against that backdrop, assurances about the water usage of closed-loop cooling systems really don't cut through. And, though developers will doubtless get cannier at this, cash payoffs to communities can easily look like subterfuge and do little to help.
3/ Still, moratoriums are at this stage much more bark than bite (h/t @mayabarkin for the data there). Only about 2GW seems actually delayed. The bigger risk is deterring newer projects, but I'd still guess that affects the exact locations more than whether the build-out happens at all.
4/ The question I'm left with is what comes after data centres as the big AI political issue into next year. A few obvious candidates are: a big cyberattack, the sheer scale of AI-related PAC spending in the 2028 cycle, the early innings of job displacement.
5/ What the data-centre issue should tell us, I think, is that voters' hostility to new technologies that change their lives in unpleasant ways is vast—especially when those changes are less positive-sum than data centres. So, however mad you expect AI as a technology to make the next few years, scale that up accordingly to account for a much more scrambled politics.
Do take a look at my piece on this here: economist.com/united-states/…
And at our Checks and Balances podcast, where I contributed a short diary featuring lots of conversations with voters economist.com/podcasts/2026/…
Finally, a datacenter the neighbors can live with: rooftop solar, its own wind turbine, zero water draw, flower beds for the planning commission, and the whole thing fits on a folding table. 0 MW of IT load, but you can’t have everything.
Maya Barkin retweeted
Our Moratoriums piece led by the DC Queen @mayabarkin talks about how it's entirely inaccurate that the rise in political backlash would somehow compromise the AI infra buildout.
Our DCM Model forecasts +38 GW of Datacenter IT capacity delivered in the US in 2027, more than double that of 2026, and we believe that less than 10% is delayed due to local moratoriums and the New York’s executive order.
The main issue here is the fact that many analysis of moratoriums out there haven't went into the granular-level of details, and it requires a parcel-level, project-by-project analysis, which SemiAnalysis did effectively.
And, since the general DC buildout sentiment on the federal level is positive, we see little threat from moratoriums.
Maya Barkin retweeted
The datacenter moratorium curve is rising. But where does the risk actually sit? (1/5)🧵
Maya Barkin retweeted
What's happening in Germany is a political apocalypse. In Mecklenburg-Vorpommern, AfD racks up its second win. In Berlin, the successor party to the East German communists becomes the strongest party. Watch my live stream from earlier today on all this:
robinjbrooks.substack.com/p/…
CDU losses in two key states yesterday put German Chancellor Merz on shakier ground and make growth reforms in Germany less likely to pass. The far-right party AfD topped the polls in Mecklenburg-Vorpommern and anti-capitalist die Linke won in Berlin.
Good Morning from Germany, where voters have delivered a historic blow to the political mainstream. Merz’s CDU has been shut out of a state parliament for the first time in the Federal Republic’s history: just 4.9% in Mecklenburg-Western Pomerania, where the far-right AfD won 38.2%. In Berlin, The anti-capitalist Left topped the poll w/25.7%. As Germany’s electorate polarises, pressure on Chancellor Merz is mounting and so is the risk that much-needed growth reforms are watered down or blocked altogether.
Maya Barkin retweeted
Good Morning from Germany, where voters have delivered a historic blow to the political mainstream. Merz’s CDU has been shut out of a state parliament for the first time in the Federal Republic’s history: just 4.9% in Mecklenburg-Western Pomerania, where the far-right AfD won 38.2%. In Berlin, The anti-capitalist Left topped the poll w/25.7%. As Germany’s electorate polarises, pressure on Chancellor Merz is mounting and so is the risk that much-needed growth reforms are watered down or blocked altogether.