In a few years, every computer and every robot could become a micro Data Center.

Rumor has it that Apple is developing the M7 Ultra, which can support up to 1.5TB of unified memory. While the specifications and launch timeline have not yet been confirmed, the trend is already very clear:
On-device machines will only keep getting more powerful, while models will become smaller and more efficient through quantization, distillation, and sparsification.
Models that today require a Data Center to run may, in a few years, run on a single Mac Studio, AI PC, or even a robot.
In the future, large Data Centers will still be responsible for training Frontier Model; but for day-to-day AI Agent work, image generation, data analysis, and robot decision-making, there will be more and more tasks completed directly on-device.
Especially once robots become widely adopted, each robot will need powerful chip processing for vision, speech, and actions, but they won’t work 24 hours a day.
When charging or on standby, this idle compute can be released to handle split-up tasks such as batch inference, model evaluation, and video generation—then earn income for its owner.
Of course, 10,000 distributed computers do not equal a 10,000-GPU Cluster. They aren’t suitable for jointly training the same super-large model, but they can simultaneously complete 10,000 different inference tasks.
So Data Centers won’t disappear, but compute power won’t always be concentrated only in Data Centers.
The last era was one where every household could get online.
In the next era, it may be that every household can provide compute power, so that when your computers and robots are idle, they continue to work for you and create passive income.