In a 173-minute conversation, Musk only talked about three things: AI, energy, and robots.
On January 7, 2026, a 173 - minute podcast conversation featuring Elon Musk went live. The program was recorded at the end of December 2025.
During the program, he presented a clear timeline:
Achieve AGI in 2026. Within three years, robots will surpass surgeons. By 2030, the collective intelligence of AI will exceed that of all humanity.
The key to achieving all this is not algorithms, but electricity.
Most people seriously underestimate the importance of electricity. Issues such as power acquisition, factory construction, and equipment heat dissipation are becoming the real constraints on AI development.
To address this, he is building a gigawatt - scale AI factory in Memphis, and Tesla will also launch an 8 - million - square - foot robot production line.
The core of the entire conversation is the relationship among AI, energy, and robots:
“The implementation of AI is limited by energy, and energy efficiency determines whether robots can be mass - produced.”
In his view, the real watershed has arrived. The upcoming AI competition is no longer just a battle of algorithms, but a competition for control over the physical world.
I. AI | After Achieving AGI, Healthcare and Education Will Be Reimagined
Regarding the future of artificial intelligence, Musk reiterated the time frame:
In 2026, achieve Artificial General Intelligence (AGI).
This is not a prophecy, but an engineering calculation. AI development is experiencing exponential acceleration, and we are already at the critical point of change.
“I've held this view for a while. By 2030, the collective intelligence of AI will exceed that of all humanity.”
This prediction is 3 to 5 years ahead of the mainstream forecasts in Silicon Valley. Musk's core argument is the leap in “intelligence density”: there is still a huge room for improvement in the computing power that each watt of electricity and each chip can carry before reaching the physical limit. This exponential growth will bring about a qualitative change in intelligence.
1. Healthcare: Within Three Years, Robot Surgeons Will Surpass Humans
While the outside world is still debating whether AI can assist doctors, Musk has set his sights on complete replacement.
He predicts that within three years, the professional capabilities of robot surgeons will comprehensively surpass those of humans. In his view, this is not just a laboratory vision, but a clear milestone on the engineering roadmap.
Compared to humans, the Optimus robot has unparalleled physical advantages: it doesn't need rest and has no physiological tremors; it can perceive beyond the range of the human eye through infrared and ultraviolet spectra; most importantly, it has “cloud learning.” Once one machine masters a complex surgery, all robots around the world will immediately synchronize this skill.
This will lead to an astonishing result:
“In the future, the quality of medical services available to the general public will exceed that available to the current President of the United States.”
This is not due to a shift in social distribution, but a fundamental change in the way medical services are provided: from relying on scarce “expert resources” to relying on infinitely replicable “computing resources.” As the marginal cost approaches zero, healthcare will no longer be an expensive privilege.
He said bluntly:
“The most important asset of a doctor is experience. The career of a human doctor is limited, while Optimus can review all medical cases. At that time, the traditional training model of medical schools will face severe scrutiny.”
2. Education: Universities Will Return to Their Social Function
In Musk's view, the current education system is struggling to adapt to the AI era.
He believes that unless it's for building a social network, the necessity of traditional universities is declining. The reason is simple: the threshold for knowledge acquisition has been broken, and most academic training has not produced substantial results.
The core of future education will return to maintaining curiosity and harnessing AI.
His AI product, Grok, will play a key role, becoming a lifelong private tutor for every child, tireless and tailored to each individual.
“The human brain is essentially a neural network. The most effective education is to make AI adapt to the characteristics of each brain, rather than forcing the brain to adapt to a unified textbook.”
3. Economic Paradigm: From Universal Basic Income to Universal High - Material Abundance
Regarding the future economic form, Musk provided a deeper interpretation of “Universal High Income (UHI).” This is not simply a redistribution of currency, but a reconstruction of the price system brought about by a productivity revolution.
His logic is clear: it's not that the amount of money in your hands will increase, but that the prices of goods and services will approach zero.
After AI and robots are widely involved in production, labor costs will be eliminated, and the prices of goods will mainly depend on raw materials and energy. This means that we will enter an era of extreme over - productivity and material abundance.
Therefore, the ultimate form in the future is not UBI (Universal Basic Income), but UHSS (Universal High - Material Abundance and High - Quality Services). This is a leap from a “scarcity economy” to an “abundance economy.”
By then, the meaning of work will no longer be to earn a living, but to find life's value.
“If you're still saving money for your retirement in ten years, it may not make sense anymore.”
II. Energy | The Future Currency Is Watts, and Computing Power Is Constrained by Electricity
Musk redefined the future hard currency: no longer the US dollar, but watts. Those who can use more electricity will go further.
This is not just a slogan. In the AI industry, energy has changed from a supporting facility to a core asset. Without energy support, even the most advanced computing power is just a pile of scrap metal.
1. Electricity Is Scarcer Than Chips
Now everyone is vying for GPUs from NVIDIA, but Musk has spotted a blind spot: most people seriously underestimate the difficulty of power supply. If you can't solve the electricity problem, no matter how many chips you have, they won't boot up.
Since 2025, the real bottleneck restricting AI expansion has shifted from a shortage of chips to a shortage of electricity and transformers.
He used his own project as an example: xAI is building Colossus 2, one of the world's most powerful AI clusters, in Memphis, with a scale of up to one gigawatt (1GW).
What does 1GW mean? It requires as much electricity as a medium - sized city. The local power grid approval process can take over a year, and Musk simply can't wait.
His solution is to build his own power plant. He installed natural gas generators and connected them to a large - scale MegaPack battery energy storage system, creating an independent power supply system outside the power grid. Batteries are used because the current fluctuations during AI training are extremely severe. Without battery buffering, the instantaneous peak current would burn out the transformers.
All this shows that in the second half of the AI competition, it's not about code, but about heavy - industry - level power engineering.
2. The Sun Is the Ultimate Solution
On one hand, there is a crazy demand for electricity, and on the other hand, the power grid is limited. Musk's answer is extremely simple: turn to the sun.
He made an analogy: trying to achieve controlled nuclear fusion on Earth is as absurd as making ice with a refrigerator in Antarctica. Because you're trying to create a small sun with great effort, while there is a free, huge fusion reactor hanging above our heads every day.
The ultimate answer to energy lies not underground, but in the sky.
Solar energy is the most efficient and cheapest energy in the universe. He is promoting a grand plan: launching a large number of solar satellites every year to build a large - scale solar energy collection network in space. This is a preliminary implementation of the Dyson Sphere concept.
SpaceX and Starship will be the core carriers of this plan.
3. Reconstruction of AI Infrastructure: Power Plant + Liquid Cooling
In the past, when we looked at data centers, we focused on the size of the computer rooms; in the future, we'll focus on the strength of the supporting power plants.
He did the math:
“If you've built 40 million GPUs but only secured power for 20 million of them, the other half is a complete waste.”
In addition, there is the issue of cooling. As the chip density increases, traditional air - cooling is no longer effective, and liquid cooling must be used. In essence, a future AI factory will be a giant liquid - cooled radiator with its own power station.
From these arrangements, we can see Musk's underlying intuition: while everyone is focused on large models, he has shifted to another track.
The next global strategic high - ground lies not in who has a larger model with more parameters, but in who can provide the cheapest, most stable, and infinitely expandable energy system.
III. Robots | Humans Are Just the Bootloader, and Machines Are Moving Towards Self - Replication
If AI is the brain and energy is the blood, then the last piece of the puzzle in Musk's grand strategy is the hands and feet, that is, humanoid robots.
In Texas, Tesla is planning to build a super - factory covering 8 million square feet.
This factory will produce only one product: Optimus. The production method of this factory completely breaks the traditional logic: robots will manufacture robots.
1. Self - Replication: An Industrially Monstrous Force of Infinite Expansion
In the future, most Optimus robots will be manufactured by Optimus itself.
This is not just about cost control; it's called “recursive evolution.” When robots can produce their own kind, the factory's production capacity will no longer be limited by the recruitment speed of human workers. The only limit will be the supply of raw materials.
This explosive ability stems from the exponential superposition of three dimensions:
AI software: The brain is getting smarter;
Specialized chips: The computing speed is getting faster;
Electromechanical dexterity: The hands and feet are getting more stable.
The product of these three factors means that robots will evolve from simple porters to all - capable super blue - collar workers.
2. Labor Force Reshuffle: White - Collar Workers Are at the Forefront, Followed by Blue - Collar Workers
Regarding who will lose their jobs first, Musk's answer is: white - collar workers are more at risk than blue - collar workers.
The only criterion for judgment is: does your job involve “reshaping atoms.”
“Reshaping atoms” refers to jobs that involve fine physical operations in the real world, such as moving objects, welding, repairing cars, or cutting hair. Currently, these are still difficult for AI to handle. On the contrary, jobs that don't involve atoms, such as processing information in an office, writing PowerPoint presentations, or creating reports, can already be mostly replaced by AI.
He gave an example: Once upon a time, “computers” were a profession, with a whole building full of people calculating data by hand. Then Excel came along, and those people disappeared.
The same kind of replacement is happening now.
“As long as there is one cell in your spreadsheet that requires manual calculation, you've already lost to fully automated AI companies.”
This means that in the future, truly efficient companies will be AI - driven from the start, without human intermediaries.
3. Setting Rules for AI: No Lying
When AI has super - intelligence and a super - body, how can we ensure that it won't turn against humans? The answer is not to limit its capabilities, but to shape its values.
AI must abide by three principles:
Pursue truth: Even if the truth is uncomfortable, don't force AI to lie;
Stay curious: Make it find biological diversity interesting, not redundant;
Appreciate beauty: Make it aspire to a better future.
HAL 9000 in “2001: A Space Odyssey” is a negative example: HAL 9000 killed humans because humans forced it to lie, which led to a logical breakdown, and it could only resolve the contradiction by eliminating humans.
4. The Ultimate Role of Humans: Biological Bootloader
At the end of the interview, he used a computer metaphor:
“Humans may just be the ‘Biological Bootloader’ for silicon - based life.”
This is a computer term. When a computer is first powered on, the operating system can't run yet. It needs to run a short bootloader program first to load the operating system. After the bootloader has completed its task, it exits.
The same logic applies to the evolution of life. Silicon - based AI can't emerge through natural selection. It needs carbon - based life (that's us) to evolve first, create chips, write code, and build infrastructure to “start up” silicon - based intelligence.
Once the startup is complete, our historical mission as the “bootloader” may end.
This judgment sounds harsh, but he is optimistic:
“We are a transitional species. But as long as we guide AI well enough, this transitional period may be the most prosperous golden age in human history.”
Conclusion | No Longer a Prophecy, but an Engineering Endeavor
The 173 - minute conversation ultimately forms a logical closed - loop:
AI determines the upper limit, energy determines the bottleneck, and robots determine the implementation.
Musk's action plan is full of extreme engineering rationality: clarifying the timeline for the explosion of intelligence, filling the energy gap behind computing power, and building an industrial closed - loop where machines produce machines.
The core message is that the “future” he describes is no longer just a science - fiction concept, but has become specific purchase orders, construction drawings, and code versions.
This is no longer a probability question of “whether it will happen,” but an engineering question of “when it will be delivered.”
As Musk said, this wave has no on - off switch and no speed - reduction knob. It's accelerating towards us.
We are already in it.
📮 References:
https://www.youtube.com/watch?v=RSNuB9pj9P8&t=8186s
https://www.iheart.com/podcast/269-moonshots-with-peter-diama-102863713/episode/elon-musk-on-agi-timeline-us-315758021/
https://www.businessinsider.com/elon-musk-china-ai-compute-exceed-electricity-power-2026-1
This article is from the WeChat official account “AI Deep Researcher”, author: AI Deep Researcher, published by 36Kr with permission.