Hyperdimensional
hyperdimensional.co · Personal blog · on HN since 2024
Hyperdimensional is a newsletter on AI policy and the AI industry, covering US and Chinese strategy, regulation like SB 1047, and AI agents.
What HN saysThe only HN comment given is about solar power costs and doesn't discuss the blog; HN links it for AI policy analysis.
27HN people shared it
1linked it in comments
45stories submitted
138story points
4 months agolast shared
2024people per year2026
Most-shared posts
- 1America Is All-In on Deep Learning; China Emphasises Robotics and Hardware4 people · 10 points · first shared 2025 · 2 comments on HN
- 2On Recursive Self-Improvement4 people · 6 points · first shared 2026 · discussion on HN
- 3Where We Are Headed3 people · 18 points · first shared 2025 · 5 comments on HN
- 4Anthropic and the Department of War3 people · 5 points · first shared 2026 · discussion on HN
- 5Among the Agents3 people · 3 points · first shared 2026 · discussion on HN
- 6Leviathan Waking2 people · 3 points · first shared 2026 · discussion on HN
- 72023 Or, Why I am Not a Doomer2 people · 2 points · first shared 2026 · discussion on HN
- 8California Senate Passes SB 10471 people · 44 points · first shared 2024 · 42 comments on HN
- 9Why Manus Matters1 people · 6 points · first shared 2025 · discussion on HN
- 10Tough Rocks: Eliminating the Chinese Rare Earth Chokepoint1 people · 4 points · first shared 2025 · discussion on HN
- 11Is AI in Trouble at the Supreme Court? (Part One)1 people · 3 points · first shared 2024 · discussion on HN
- 12Heiliger Dankgesang: Reflections on Claude Opus 4.51 people · 3 points · first shared 2025 · discussion on HN
- 13Texas Might Regulate AI1 people · 3 points · first shared 2025 · 3 comments on HN
- 14An Extraordinary Alien: OpenAI's O1 Challenges AI Policy1 people · 2 points · first shared 2024 · discussion on HN
- 15The Timing of AI Regulation1 people · 2 points · first shared 2024 · discussion on HN
What HN people wrote when they linked it
wrsh07 · 2025-12-13
Casey Handmer is a huge solar bull and his estimate is that solar becomes cheaper than any other form of electricity even when generated from northern states by 2030 (likely sooner)
Iirc solar is meaningfully more efficient (30-50%) in southern states, so it will likely make sense to place energy intensive workloads in locations with more direct sun.
However, the cost of transmitting additional power is interesting and complex. Building out the grid (which runs close to capacity by some metric^) is expensive: transmission lines, transformers or substations, and acquiring land is obvious stuff. Plus the overhead of administration which is significant.
So there's a lot of new behind-the-meter generation (ie electricity that never touches the grid)^^
With all that in mind, I expect energy intensive things will move south (if they have no other constraints. Eg cooling for data centers might be cheaper in northern climes. Some processing will make sense close to where materials are available) But a significant amount of new solar will still be used in northern states because it's going to be extremely cheap to build additional capacity. Especially capacity that is behind-the-meter.
^ but not others! Eg if you're willing to discuss tradeoffs you might find dozens of gw available most of the time https://www.hyperdimensional.co/p/out-of-thin-air
^^ patio11 has a good podcast about this https://www.complexsystemspodcast.com/episodes/the-ai-energy-bottlenec…