Most likely - fusion ignition in the 2030s. Fusion supplying 10% of global electricity in the 2050s.
https://www.metaculus.com/questions/?order_by=-rank&main-fee...
There is not obviously lots of money to be made in rapidly building out fusion power plants.
https://en.wikipedia.org/wiki/DEMOnstration_Power_Plant#Time...
ITER is doing that NOW and NONE of headline-generating startups have even started...
So it seems to me that ITER is both contributing more engineering insights and closer towards a full-fledged fusion power-plant, and unlikely to be overtaken in these regards anytime soon, if ever.
We have superconductors now that can support much stronger magnetic fields than ITER uses. That lets us build smaller tokamaks with the same output as ITER. Several startups are using them, led by MIT spinoff CFS. Their SPARC reactor should have output similar to ITER's, in a reactor half the size of JET, which was built in a year. They expect to attempt net power in 2025 and a lot of researchers think they'll succeed.
Assuming that works, as similar reactor the size of JET gets them to commercial output levels.
So even IF the whole SPARC thing goes according to play, it's going to be miles away from anything resembling a competitive power plant, while ITER is potentially something like a quarter way there (assuming that operating at 2GW thermal begins being somewhat attractive/feasible for a fusion-powerplant).
Personally, I strongly doubt that fusion power is EVER going to be an option because of simple economics; I just don't really see how a vacuum chamber surrounded by superconducting magnets, cooling systems, turbines and turbogenerators is ever going to compete (financially) with just slapping PV panels on a roof, putting some battery banks in the basement, hooking it up with an inverter and just repeating as necessary until power demands are met...
A bigger research reactor is simply better in that it enables you to investigate/solve problems that are related to scale, and all those MUST be solved before a commercial plant can be built.
Until ARC or an even further removed successor catches up to ITER in size/power, the SPARC project is just yet another toy reactor IMO.
https://www.gem.wiki/Existing_U.S._Coal_Plants#Size_comparis...
Coal plants of course use steam turbines:
https://www.tva.com/energy/our-power-system/coal/how-a-coal-...
Because your chart cements my point: Power produced by coal plants smaller than ITER (<500MW thermal output, or <250MW electrical power in your chart) is negligible because building those makes evidently no sense economically (almost all installed capacity is in big plants).
And thats with coal where building smaller actually reduces operating cost from fuel (unlike fusion) plus needs no vacuum chamber, cryocooling or dealing with neutron activation...
Getting D-T fusion competitive will be a challenge for everyone, but SPARC/ARC has an obvious advantage over ITER/DEMO by having way, way lower capital costs.
This isn't entirely ITER's fault. The superconductor technology SPARC is using didn't exist when ITER was designed.
Achieving stable fusion does not make practical power from it possible. It would necessarily cost >10x fission, and fission is not competitive.
We can be absolutely ,confident that by 2100, there will have been zero kWh of commercial power from Tokamak fusion. It is just barely possible that D-3He fusion might work by then, but it would still struggle to compete. It might find use in outer-solar-system spacecraft.