~6m1:25:06
The Economist

The full-length interview with Elon Musk | The Economist

Jul 29, 2026

Read: ~7m · You save: 78 min

The Full-Length Interview with Elon Musk | The Economist

Elon Musk predicts AI will surpass all human intelligence within five years. He discusses abundance, jobs, AI safety, China, and his controversial politics in this shar

Interview transcript recast as a published article. Date of record: 2026-07-29.

Introduction

In an interview with The Economist conducted at Tesla Giga, Elon Musk addressed the future of artificial intelligence (AI) and robotics, the economics of an AI-driven economy, the safety of advanced AI, the geopolitics of AI and space infrastructure, and his involvement in politics. The interview covered his predictions for 2036, his views on AI risk and regulation, and his positions on several political and policy questions.

Predictions for 2036

Musk said that AI could exceed the sum of human intelligence in roughly five years. He estimated that within 10 years, AI would be "far greater" than all human intelligence combined. He stated that AI would then be able to perform any task better than humans, apart from being human.

He described the most likely outcome as "an age of amazing abundance," in which anyone could have anything they can imagine. He identified a major global thermonuclear war as a possible disruption to this outcome.

Economic Model: Digital and Physical Intelligence

Musk described the economy as the production of goods and the provision of services. In his view, the economy has two components: digital intelligence and physical intelligence.

  • Digital intelligence: AI systems that advance rapidly but remain largely confined to the digital realm.
  • Physical intelligence: Humanoid robots that serve as "end effectors," allowing AI to act on the physical world. Musk said that the economy will require "lots of bots."

Musk argued that combining large numbers of robots with large amounts of digital intelligence would produce a "quasi-infinite" economy. He said no existing analogy captures the scale of the change.

Money in a Post-Scarcity Economy

Musk predicted that money "won't matter" in 2036. He argued that people use money to obtain goods and services. If robots and AI produce more goods and services than any human can consume, he said, money would lose relevance.

He also predicted that deflation, rather than inflation, would become the main monetary risk. In his explanation, inflation reflects the ratio of money to goods and services. If output of goods and services grows faster than the money supply, prices fall. He said the Treasury could issue payments to citizens without causing inflation under this condition, because the money supply would grow more slowly than output.

Work

Musk said work would become "optional." He compared future work to gardening: people would still grow vegetables, but the output would be less uniform than commercially produced goods.

Artificial Intelligence and Human Control

Asked whether humans would remain in control, Musk said that it is "unlikely" that they would. He compared the relationship to that between humans and chimpanzees. He said that a large intelligence gap makes it "hard to imagine" that the less intelligent party would be in charge.

He stated that the priority for AI safety is for AI to have good values, to care about humanity, and to want humans to be happy and prosperous. He said his "biological neural net" indicated that the most important safety property is to be "maximally truth seeking and curious." He stated that this property would foster humanity.

AI Risk

Change in Musk's Position

Musk said that AI and robots still carry risk, and that the risk is "not zero." He said he has reached the philosophical conclusion to "look on the bright side." He stated that he cannot see a way to stop AI progress, and that even if a stop button existed, humanity would probably not press it, because the most likely outcome is "incredible abundance for all."

The interviewer noted that Musk had previously:

  • signed a 2023 letter calling for a slowdown in AI development, which Musk described as one of roughly 500 signatories;
  • stated on a podcast, approximately ten years earlier, that rapid recursive self-improvement in AI was the thing that most terrified him, and that humans would be "the pet Labradors at the best";
  • estimated, in a podcast appearance the previous year, a 10 to 20 percent chance that "killer robots" would wipe out humanity.

Musk did not dispute these statements. He said the 10 to 20 percent estimate was not something he would revise now, and he described the end of the universe under the standard model of physics as a "heat death" in which all changes end. He said that if this outcome holds, "it really is all about the journey."

Acceleration of AI by Musk-Associated Organizations

Musk said that he co-founded OpenAI to act as a counterweight to Google, which at the time had a near-monopoly on AI. He said that Anthropic later spun out of OpenAI and that Anthropic is now a leader in AI. He stated that his actions had produced knock-on effects that accelerated AI, which he said was not his intention. He concluded that "all roads lead to acceleration of AI."

Safety Proposals and Industry Coordination

Musk described a proposal by Demis Hassabis, chief executive of Google DeepMind, for a private-public regulator, which Musk called a good starting point for discussion. Musk said he had spoken with Hassabis for a few hours before the proposal was published.

Musk's recommended immediate measures:

  1. Leading AI companies meet, or hold a call, once every few weeks to discuss safety and security issues.
  2. Before a new frontier model is released, competing AI companies test that model for harmful effects. Musk said a review period of one to two weeks would suffice, and that testing could take place through an API, without handing over the model.
  3. Companies recommend pausing a release when a model presents security issues.

Musk said that competitors are better placed than government officials to judge whether a model is safe, because officials often lack deep technical knowledge. He said competitors have an incentive to keep one another "honest."

He said he had proposed the arrangement to Hassabis and would raise it with Dario Amodei, chief executive of Anthropic. When asked why the leading companies had not already coordinated, Musk said the proposal had not yet been put into practice.

Musk cited the Motion Picture Association as a model. He described it as an industry group that decides movie ratings for audiences, and said the arrangement "seems to have worked quite well."

Role of Governments

Musk said that if a leading company judges a model to be very dangerous and the developer does not take steps to address the danger, the company should alert the government. He said that only the US and Chinese governments have the power to act in such cases. He said the Chinese government should be included in safety arrangements, along with Chinese companies developing frontier AI models.

The interviewer noted that the US government had used the threat of export controls to block the wide release of Anthropic's model, Mythos, over cybersecurity concerns. Musk said that the government did not identify the risk independently. He said that Amazon identified the cybersecurity risk and contacted the White House.

Musk said he thinks governments should hold this power. He said the government should act if a company does not address a known risk, and that this is "literally what just occurred" with Mythos.

Musk said action should be immediate. He said the pace of model development is such that a system should exist well before six months elapse.

Competition with China

Model Capability

Musk said Anthropic's Mythos model, from which the Fable model is derived, was ready in February. He said Anthropic has a more advanced model than Mythos, and that the company could release it at any time.

He said Fable is "clearly the smartest model" currently available. He said the Chinese model Kimi K3, released the previous week, is close to Fable in capability.

He said Chinese AI companies perform well with comparatively little compute. He said that if they gained substantial compute, they would likely lead.

Constraints

Musk identified the constraints on AI as electricity and AI chips. For physical AI, he identified robots as an additional constraint. He said:

  • China has more electricity than the United States, and more than the United States, Europe, and India combined.
  • He estimated that China will reach four times the electricity production of the United States, roughly in proportion to population.
  • Inside China, chips are the binding constraint, because the US has banned export of its latest AI chips to China. Chinese firms are becoming more efficient with the chips they have.
  • Outside China, electricity is the binding constraint. Musk said the rate at which AI chips are produced exceeds the rate at which new electricity comes online.
  • Musk said China is "closer than most people realize" to solving the lithography problem, which would enable production of AI chips in very large volumes.
  • Once power constraints are addressed through AI data centers in space, chips will again be the constraint outside China.

Musk said the water use of AI is negligible, while its electricity demand is very high.

US Policy

Asked whether the US should ban use of Chinese models such as Kimi K3, Musk said a US law could prevent US companies from using them, but could not stop the rest of the world from doing so. He said such a ban would not stop China from leading in AI.

Musk said Chinese robotics firms are strong. He referred to robot sporting events in China, including robot boxing and combat matches.

Politics and Public Concern

Job Displacement

The interviewer noted that roughly 75 percent of Americans are worried about AI, and that Amodei had said that about 50 percent of entry-level white-collar jobs could disappear within a few years. Musk said that Amodei "did dig his own grave" by describing Mythos as frightening while also announcing its release. Musk said that Amodei would probably choose his words differently in future.

Musk said AI will be able to do any job better than any person can do it, starting with digital work and any task performed at a computer or phone. He said:

  • AI already surpasses at least 90 percent of professional software engineers at writing software.
  • The ability will soon reach 99 percent, and then reach a level he called "Stockfish level," at which competition becomes impossible.
  • He said AI would be better than humans at everything in the same way Stockfish is better than any chess player, including Magnus Carlsen.

He said the transition would be "a bumpy road."

Historical Comparison

Musk said that many jobs have disappeared because of computers. He cited "computer" as a job title for people who performed calculations, including those who calculated bank interest in large office buildings. He said some workers resisted the loss of those roles, but few people now want them. He said the difference now is the pace of change.

Musk recommended Iain M. Banks's Culture novels as a depiction of an AI future. He said he does not find the future they describe appealing because humans have little agency compared with the minds that run it. He also said that Banks was a socialist with a different political philosophy from Musk's.

Redistribution

The interviewer asked whether high taxes on capital, or large redistribution, would be needed to fund universal income. Musk said the Treasury should "simply issue people checks." When the interviewer noted that issuing checks could cause inflation, Musk repeated his argument that output growth would prevent inflation.

Musk said he pays a large share of tax. He stated that he set a record for the most taxes paid by an individual. He said he pays about 40 percent federal income tax on his stock options, and about 15 percent California state tax, because he spends roughly 30 percent of his time in California. He estimated his total tax at about 45 percent. He said his estate would be taxed at a similar rate after his death.

He said he expects to pay "many trillions" in tax.

Public Share Ownership

Musk said he took SpaceX public so that the public could own a share of the company. He said a private company is limited to a few thousand shareholders, while a public company can have millions. He said he is not aware of plans for shares to go to ordinary Americans, though he acknowledged discussion about "Trump accounts" as a possible vehicle.

Concentration of Control

The interviewer said that Musk holds about 80 percent of voting shares after the SpaceX IPO and can be removed only by a vote of shares he controls. Musk said that Alphabet is controlled by Larry Page and Sergey Brin, and that Mark Zuckerberg has voting control at Meta. He said his control is more extreme in degree but similar in kind.

Musk said he needs control to focus on long-term investments with five-to-ten-year horizons. He said public markets pressure companies to deliver quarterly results. He cited SpaceX's plans for a Moon base and Mars base, which the S-1 prospectus lists, as examples of expenditure that would depress short-term earnings. He said portfolio managers face short evaluation periods, which drives short-term pressure on companies.

Asked about key-person risk, Musk said the companies would "do very well for several years" without him, and that SpaceX and Tesla have roadmaps. He compared his position to Steve Jobs at Apple. He said Apple has continued to make strong products, but lacks a Jobs-level breakthrough product.

Mars, Space, and Consciousness

Musk said his goal is to "extend consciousness beyond Earth" and create a growing civilization on the Moon, Mars, and elsewhere in the solar system. He said he wants to maximize the future reach of consciousness. He said he would like the future of Star Trek or Star Wars to become real. He described Star Wars as the first film he saw in a theater, at age six.

Musk said that Starship is the largest and most powerful flying object ever made, and that SpaceX intends to launch it more than once per hour. He said that Ray Kurzweil's predictions have been "remarkably accurate."

Geopolitics: Starlink and Ukraine

Musk said the Ukrainian government has publicly thanked Starlink for its role in defending Ukraine.

He said SpaceX never sold Starlink to Russians. He said Russian forces in occupied territories obtained Starlink units that had been ordered through Ukraine and smuggled into eastern Ukraine. He said some units were used in attacks.

To address this, SpaceX worked with the Ukrainian government to create a whitelist of approved Starlink terminals. Musk said the whitelist has downsides, because many innocent users lost access.

Asked whether it is appropriate for an individual to control a capability that can shape a conflict, Musk said he is "not sure" that SpaceX determines the outcome of the war. He said:

  • The border has barely moved in two to three years.
  • He hopes for a peace agreement soon.
  • Peace will require concessions, and a demand for full Russian withdrawal is "not realistic."
  • Russian conscripts are also dying at the front.

Musk said he has advocated peace for a long time and had posted that Russia must receive some concession. He said his position is not pro-Russian.

The interviewer noted Ukraine's recent drone strikes inside Russia and stated that the balance of the war appears to be shifting. Musk said Ukraine will not retake the occupied territories.

Politics in the United States

DOGE and Government Spending

Musk said he "got a little too involved in politics." He said the goal of the Department of Government Efficiency (DOGE) was to address the federal deficit, and that interest payments on the debt exceed the combined spending on the Department of War and US intelligence agencies.

He said DOGE officials asked for recipient information on certain payments. He said that payments described as going to Africa had been routed to a Deloitte & Touche account in Washington, D.C., and that officials could not obtain recipient details.

The interviewer said the abrupt shutdown of US aid programs likely caused unnecessary suffering. Musk disagreed. He said "zero" people died as a result. He said that he had received reports from Africa, and that many nongovernmental organizations (NGOs) could step in. He said the Gates Foundation has $50 billion, and that MacKenzie Scott has given away tens of billions of dollars.

The interviewer said the abrupt halt of antiretroviral funding would cause deaths if people lost access to treatment. Musk said such deaths would occur only if no alternative funding could be secured. He said USAID was a political organization, and that funding had been moved to the State Department rather than stopped. The interviewer said health programs were shut down in Africa. Musk said these claims are false.

Musk said the Trump administration is "excellent" on balance and that no administration is perfect.

Politics in Europe and the United Kingdom

Musk's Position

The interviewer said Musk frequently supports populist and far-right parties in Europe. Musk said he supports "normal people." He said policies he advocates, including secure borders, safe cities, and sensible government spending, were mainstream 10 to 15 years ago. He said that a speech by Barack Obama or Hillary Clinton would now be considered far right by some audiences.

Musk said that his political views are centrist and that describing them as far right is false and misleading.

Crime and Migration

Musk said that the UK is not as safe as it could be and that "civil war in Britain is inevitable," if current trends continue. He said this would occur if a large and growing group whose beliefs are "antithetical to Western beliefs" tried to impose those beliefs.

The interviewer said that London is much safer than any US city. The interviewer said violent crime in the UK is declining. Musk said he lives in the UK part of the time and disputed the interviewer's claims about crime rates. He said he had last visited the UK a few years earlier, and that he spends time there regularly.

On crime, Musk said violent crime in the US is driven by repeat violent offenders who are not incarcerated, and by the availability of guns. He cited the number of heat-related deaths in Europe as higher than gun deaths in the US.

Musk said that he is against "rape and murder" and against the imposition of laws contrary to Western norms. He said migration should be limited to people who will be productive members of society, will be honest, and will not mistreat others in the host country.

The interviewer said Europe's welfare states provide benefits that create an economic incentive for migration. Musk agreed. The interviewer said migration also brings benefits to an aging continent, and that controlled immigration is beneficial. Musk said he is in favor of immigration under the conditions he described.

Free Speech

Musk said that the UK has jailed individuals for social media posts, and that he has "written editorials against it." The interviewer said that restrictions on speech in the UK are problematic in some areas.

The interviewer criticized a 48-hour citizen vigilante film, which Musk had promoted on X. The interviewer said the film portrays Europe as dystopian and glorifies vigilante violence against a Muslim immigrant family. Musk said the film is "worth watching."

Statements on Parties

The interviewer said that Musk's support for some European parties is harmful. The interviewer said that some members of the Alternative for Germany (AfD) party are neo-Nazis, and that Musk's support for the party gives him influence in Europe. Musk said he hopes he is having a large impact.

Media Perception

Musk said that the favorable view of journalists in the United States is around 15 percent. He said journalists are more disliked than he is. The interviewer said that trust in journalism exists and that journalists have made mistakes.

Views on Liberalism

The interviewer said that Musk had been regarded as a classical liberal and that his posts raise doubts. Musk said he believes in secure borders, safe cities, and sensible spending, and that these views are not right-wing.

Asked which region scores better on these criteria, Musk said the US was better on borders, and that Europe was better on the other two. He said he disagreed with the interviewer's view that Europe is a "hellhole," and said he had not described it that way.

Musk's Assessment of His Influence

Musk said that his posts reach around 250 million followers on X. He said that a "Cassandra effect" makes people doubt predictions he says are accurate. He said his prediction record is strong, though not perfect.

Closing Positions on AI and Timeline

Musk said AI and robots will "dominate everything" on a time scale of less than 10 years. He said that the AI and robotics revolution will occur before civil war in Britain. He placed the timing of civil war in Britain at about 20 years away, and the AI robot singularity at about 10 years away.

Musk stated that digital superintelligence is called the singularity, as a black hole is, because the outcome after it cannot be known.

Introduction and Vision of the Future

Musk describes an AI and robotics future, predicting AI will surpass combined human intelligence in about 5 years and that an age of abundance is likely by 2036 absent catastrophe.

  • Musk predicts AI may exceed the sum of human intelligence in about 5 years
  • He expects an age of amazing abundance by 2036 if there is no World War III
  • He says the 10-year outlook is one of optimism and excitement
  • He lists his achievements: electric vehicles, space travel, AI model, humanoid robots

Making Money and the Economy of Robots

Musk argues the economy is digital and physical intelligence; AI needs humanoid robots ('end effectors') to shape atoms, creating a quasi-infinite economy.

  • AI is currently confined mostly to the digital realm and needs end effectors such as humanoid robots
  • Musk defines an economy as production of goods and provision of services
  • With vast numbers of robots and digital intelligence, the economy becomes quasi-infinite
  • He says money may not matter in 2036 because goods and services will be abundant

AI Safety, Control, and Existential Risk

Musk says humans will be unlikely to control superintelligent AI, compares this to chimpanzees, and argues AI should have good values; he signed a 2023 call for a slowdown but says stopping AI is not feasible.

  • Musk expects it is unlikely humans will remain in control, comparing to chimpanzees and AI
  • He says the most important thing for AI safety is being maximally truth-seeking and curious
  • He signed a 2023 AI slowdown letter as one of about 500 signatories
  • He earlier said there was a 10 to 20% chance of killer robots wiping out humanity, and still says the risk is not zero
  • He says the heat death of the universe makes the journey the important part

Industry Coordination and Safety Testing

Musk says his founding of OpenAI and the spinout of Anthropic accelerated AI, and he proposes a regular call among leading labs and a one-to-two-week private testing window before frontier releases.

  • Musk created OpenAI as a counterweight to Google's AI monopoly; it accelerated AI, which he did not intend
  • Anthropic spun out of OpenAI and is now a leader in AI
  • He recommends leading AI companies hold weekly or biweekly safety calls
  • He proposes competitors get a week or two of private API testing before release, to be able to recommend pausing
  • He says he talked with Demis Hassabis for a few hours before Hassabis's proposal for a private-public regulator
  • He compares the arrangement to the Motion Picture Association rating system

US-China Competition and Government Role

Musk discusses Chinese AI progress (Kimi K3 near Anthropic's Fable), electricity as the constraint, and argues US bans on Chinese models would not stop China from leading AI.

  • Musk says Kimi K3 is almost as good as Fable, which he says remains the smartest model
  • He says Anthropic has a more advanced model than Mythos, derived into Fable, that could be released at any time
  • China has more electricity than the US, Europe, and India combined and may reach four times US output
  • Outside China the constraint is electricity; inside China it is chips; China is close to solving lithography
  • He says a US ban on using Chinese models would not stop China from leading AI
  • He recommends a government step-in only when a dangerous model's company fails to address the danger

Jobs, Work, and Universal High Income

Musk argues AI will do any digital job better than humans, compares future work to optional gardening, and predicts deflation rather than inflation, while proposing the Treasury issue checks.

  • Musk says AI is already better than 90% of professional software engineers at writing software
  • He compares the coming AI capability to Stockfish level in chess
  • He says work will become optional, comparing it to gardening
  • He suggests the Treasury should simply issue checks to people
  • He predicts deflation will be the issue rather than inflation
  • He recommends science fiction, Iain M. Banks' Culture novels, and says humans have little agency there

Wealth, Taxes, and Corporate Control

Musk says he pays about 45% in taxes, explains why SpaceX went public to let the public own shares, and defends concentrated voting control to focus on 5 to 10 year goals.

  • Musk says he pays about 40% federal income tax on stock options and about 45% combined including California
  • He says SpaceX went public so the public could own shares; private firms are limited to a few thousand shareholders
  • He says he holds roughly 80% of voting shares after the IPO
  • He argues public markets pressure companies for quarterly results and he wants to invest in Mars and the Moon
  • He says Tesla and SpaceX have road maps and that Apple lacks a Steve Jobs-level breakthrough product

Starlink, Ukraine, and Geopolitical Power

Musk says Starlink was used by Russians in occupied Ukraine after smuggling, leading to a whitelist; he calls for a peace agreement and says Russia will not withdraw.

  • Starlink units were ordered through Ukraine and smuggled into occupied eastern Ukraine for Russian use, including attacks
  • Musk worked with the Ukrainian government to create a whitelist of approved Starlink terminals
  • He says the whitelist harms innocent users
  • He says a peace agreement is needed because the border has barely moved in two or three years
  • He says Russia will not withdraw and that some concession to Russia is necessary

DOGE, USAID, and US Politics

Musk defends the rapid cuts to USAID and DOGE efforts, disputes claims that cuts killed people in Africa, and says the administration is excellent on balance.

  • Musk says DOGE aimed to address a massive deficit, where interest payments exceed defense and intelligence spending
  • He says payment approvals were problematic, citing requests to wire money to a Deloitte & Touche address in Washington instead of Africa
  • He says he got too involved in politics and got carried away
  • He denies that USAID cuts caused any deaths in Africa
  • He says the Gates Foundation and Bezos could step in, and that he calls the claims 'absolute nonsense'
  • He says this administration is excellent on balance

Europe, Immigration, and Free Speech

The interview turns to Europe, with Musk calling the far right 'normal people' and claiming civil war in Britain is inevitable; the interviewer disputes his crime claims and his support for parties, and they disagree on immigration and on which is more dangerous.

  • Musk says civil war in Britain is inevitable if current trends continue and predicts it may be 20 years away
  • He claims London is safer than US cities and says he has been to the UK recently
  • He says he supports secure borders, safe cities, and sensible spending
  • Musk says he is in favor of immigration if people are productive, honest, and not mistreating others
  • The interviewer says Musk has about 240 million followers on X and criticizes his support of AfD figures
  • Musk says the AI robot singularity is 10 years away and civil war in Britain is about 20 years away