R1 illustrates the threat that computing efficiency gains pose to power generators,
We still believe data centers, reshoring, and the electrification theme will remain a tailwind,
Jevons paradox strikes again!"
As AI gets more efficient and accessible, we will see its use skyrocket, turning it into a commodity we just can't get enough of,
What we're telling people is the fundamentals are still very healthy,
Electrical grids in Europe and the US remain under-invested and one of the critical bottlenecks in terms of meeting load growth requirements,
I think it's closer to the Japanese cars in the 60s."
DeepSeek R1 is AI's Sputnik moment,
It was an over-optimism of market demand and underappreciation of the ability to do more with less,
You don't want to be carried away by the fervor of the moment,
When I say exponential, I truly mean multiple orders of magnitude: more use cases and more demand, because now, all of a sudden, it's going to be much more accessible,
If the US wants to be a leader in AI going forward, we need to make sure we'll be able to power all those data centers,
The market generally likes to do knee-jerk reactions,
It felt like the beginnings of a tulip craze,
Nobody knows if the chips are really more efficient. But it seems like there's a lot of concern,
We should have done this 10 months ago after an epic run for AMD,
Building data centres requires lots of carbon in the production of steel and also lots of carbon-intensive mining and production processes for creating the computing hardware to fill them,
In our view, DeepSeek most likely drives further demand for data centers globally with advanced computational capabilities, high capacity storage, robust networking, energy efficient designs and infrastructure,
In this instance, if DeepSeek turns out to be what everybody wants, and they sell to U.S. companies, and the U.S. companies change their algorithms to adopt to it, it just means a greater, faster broader development,
If proven true, the efficiencies used within DeepSeek's open-source model can be applied by the hyperscalers to their models, which would result in a more moderated demand,