The issue as always is economics. Current economical models means that variance and unpredictable supply chain means more expensive cost of operation. Peaker gas plants are way more expensive because they only run once in a while to cover for renewable shortfalls. If you build your factories to run on excess solar, there's going to be days where you can't operate because of rain/snowfall/sandstorm, not to mention winter. This is not a deal breaker but means that it's going to drive up costs.
The other issue is scaling up. This is terawatts of energy we're talking about here. Making and installing the solar panels/wind turbines it is one thing, actually making the factories that can use those terawatts is another. China consumes 2.5 terawatts peak, that's all of China, I can't imagine the number of factories needed to fully make use of terawatts of extra solar, or the time needed to build them. Not to mention that you still need the raw materials, you might have cheap electricity but you still need the raw ore to refine Al, water to make H2 etc. That and the labour force to man the massive amount of factories.
3rd issue is demand. We are talking terawatts of excess energy here, even if you have the factories and the raw materials to make full use of it and are churning out billions of tons of H2, hydrocarbons, aluminium etc etc, you might not have the demand for it to be economically viable, no matter how cheap it is.
Like I said, rooftop solar+wind/offshore wind+hydro+nuclear+solar farms and Twh energy storage is probably already enough to fully meet China's energy demands of 3 terawatts and more. But covering enough desert land to actually alter the climate is on another level, one of the world's biggest megaprojects, that's hundreds of thousands of Km^2 and like dozens of terawatts peak. China is probably the only country that can attempt such a project and it will still take decades.
It's possible but due to the reasons listed above, not economically viable, it will need to be a big government project running at a massive lost for a good few years/decades before they figure out what to do with all that excess electricity, if at all. But again, due to the anti-desertification effects it will probably worth it in the end, the social and environmental costs cannot be measured directly in money after all. China is already spending billions on existing anti-desertification programs that don't generate any power or direct profits. But who knows if China will actually go though with it.
It will take decades and who knows? Maybe by then automation and A.I will be good enough to find a way to economically find a way to make use of all that energy.
