"Overproduction" is nonsense, and solar panel prices have been continuously dropping dramatically for 50 years now; they're not likely to stop anytime soon.

Current solar panel production is only about a terawatt per year, which would take about 80 years to merely replace all of global marketed energy consumption (≈20TW, but those are not peak watts), and 90,000 years to reach Kardashev Type 1, where total solar power capture is roughly equal to the amount reaching the Earth. So current solar panel production is too low by roughly a factor of 1000.

We're very far from overproduction of solar panels. What we have is underconsumption of solar panels. But that has been improving dramatically year by year, and will probably continue to improve.

And if you think "it's really never been a better time to buy panels than now", you clearly haven't been watching module prices, even though I linked the historical data in the comment you are replying to, and I even quoted some prices. Prices are up 20–30% since the beginning of this year, and since November (€0.065/Wp low-cost, €0.100/Wp mainstream, €0.130/Wp high efficiency) they're up by 30–40%.

There had never been a better time to buy solar panels than last November. That's the usual situation with technological progress, actually: there has never been a better time to buy a thing than right now, and if you wait another month, it will get even cheaper. Last year, prices were mostly flat; November was the low for low-cost modules and high-efficiency modules, while mainstream panels were actually 5% higher than their low, which they had first hit in November 02024. But, still, high-efficiency panels hit record lows last year in January, August, September, and November, while low-cost panels hit record lows in July and November. The year before was one of vertiginous price declines: high-efficiency panels hit record lows in January, February, March, May, July, August, September, October, and November; mainstream panels hit record lows in January, March, June, September, and November; and low-cost panels hit record lows in January, April, August, October, and November.

This followed a three-year gradual rise in prices starting in late 02020, but we also saw record-low monthly prices in most months of 02020, most months of 02019, and all months of 02018.

In any of those months, you could have correctly said, "it's really never been a better time to buy panels than now," even though an even better time was coming up, in most cases, a month or two later. Prices have shot up this year, just like they did in 02020–02022, but within three or four years at most, they'll be lower again. Swanson's Law https://en.wikipedia.org/wiki/Swanson%27s_law remains in operation, and on the Wikipedia page you can find a nice graph from 01975 to 02024.

You said, "combine this with the global backing-off of the Paris Agreement (and subsequent subsidy defunding) and power utilities in the US abandoning clean energy goals in order to immediately increase output for proposed hyperscale facilities and you've got something of an enduring demand shortage", but this is nonsense reasoning, because you've buried a false premise: you think that solar panels are a more expensive way to generate energy than fossil fuels, which is wrong.

In fact, as should have been immediately obvious from the prices I already quoted in the comment you were responding to, solar panels are now enormously cheaper than thermal power plants, even before you add in the cost of the fuel. The proposed hyperscale datacenter facilities are precisely where the demand for today's ultra-cheap solar energy is going to come from — people in the US are spinning up portable gas turbine generators to power their data centers for regulatory reasons, but that's a desperate stopgap, not a long-term solution.

Energy is one of the largest cost items for any datacenter, so ultimately nobody can run a competitive datacenter on fossil fuels, not when they're competing against solar.

Sustainable energy without the hot air (2008)