The full-length interview with Elon Musk | The Economist

Musk opens by sketching 2036 as an age of 'amazing abundance,' where digital intelligence, already advancing through models like Grok, gets paired with humanoid robots that supply the physical labor, or 'endectors,' that pure software lacks. He describes the future economy as the fusion of digital and physical intelligence: enough robots plus enough AI compute produces what he calls a 'quasi-infinite economy,' one capable of producing more goods and services than any population could consume. Pressed on how his companies will actually make money in that world, he sidesteps with a bigger claim: money itself, he predicts, will stop mattering by 2036, because abundance will outstrip the need for it. When the interviewer notes that Tesla and SpaceX shareholders presumably expect returns, Musk concedes little ground, arguing that if robots and AI can provide more than any human could want, the entire premise of monetary value collapses.

He frames this transition as unlike anything humanity has faced before, resisting any comparison to earlier technological shifts. The only thing that could derail it, in his telling, is catastrophe on the scale of global nuclear war. Short of that, he insists the trajectory toward abundance is close to locked in, a claim he returns to repeatedly even as the interviewer keeps trying to pin down the practical, financial mechanics underneath it.

The interview's sharpest philosophical moment comes when Musk states plainly that he thinks it 'unlikely' humans will remain in control of AI, comparing the future intelligence gap between AI and humans to the gap between humans and chimpanzees, an analogy he does not treat as an insult but as a plain description of hierarchy. He is reminded that a decade ago he told Neil deGrasse Tyson that recursive self-improving AI terrified him, that humans might end up as 'the pet labradors at best,' and that as recently as Ted Cruz's podcast he cited a 10 to 20 percent chance AI could wipe out humanity. Confronted with this shift, Musk does not walk the numbers back so much as reframe his posture: he says he has arrived at a philosophical decision to 'look on the bright side' since he sees no way to actually halt the momentum of AI and robotics.

When asked whether he would board a rocket with a 10 to 20 percent chance of exploding, he answers yes, but only because in his view there is no stop button to press regardless. He admits his own actions, founding OpenAI as a counterweight to Google and inadvertently spinning off Anthropic, ended up accelerating the very race he once feared. His conclusion is almost resigned: 'you can just sort of be sad about it or join the club.'

Musk lays out a concrete, if informal, safety mechanism he discussed with Google DeepMind's Demis Hassabis before Hassabis published his own regulatory proposal: leading AI companies should hold a call every week or two, sharing early access to new frontier models so competitors can flag dangerous capabilities before public release. He likens the arrangement to the Motion Picture Association rating films, an industry body policing itself without waiting for government technical expertise it doesn't have. He argues that rival labs have every incentive to keep each other honest, since embarrassing a competitor's unsafe release serves their own competitive interest, even amid open personal hostility, he name-checks his falling out with Sam Altman over OpenAI's shift from nonprofit to an $800 billion closed-source company, and praises Dario Amodei of Anthropic as principled despite Amodei's own stumble in publicly calling a new model 'terrifying' right before releasing it.

Musk extends the idea geopolitically, arguing that Chinese frontier labs, including the Kimi K3 model he says is closing in on his own Grok, should eventually be part of the same informal safety network, since only the US and Chinese governments actually hold the power to act. He frames China's advantage bluntly: its electricity production already exceeds the US, Europe and India combined, and once chip constraints ease, he expects China to become the AI leader by sheer scale of power and manufacturing capacity.

Turning to the human fallout of automation, Musk agrees explicitly with Dario Amodei's warning that AI will eventually outperform humans at nearly every job, citing software engineering as already 90 percent solved by AI and heading toward 'Stockfish level,' a chess engine so dominant no human, however skilled, can compete. His answer to mass displacement is blunt: government should simply issue checks, funded not through redistribution in the traditional sense but through a economy so productive that, he argues, deflation rather than inflation becomes the real risk, since money creation would lag behind an exploding supply of goods and services. He resists the idea that this requires steep new taxes, noting he already expects to pay trillions in tax over his lifetime, roughly half his wealth between income and estate taxes, and suggests taxation itself becomes 'somewhat irrelevant' once money stops mattering.

On the concentration of power question, he defends his roughly 80 percent voting control of his companies as necessary to protect long-term bets, citing SpaceX's Mars and moon ambitions, from the quarterly-earnings pressure that public markets and portfolio-manager incentives impose. He compares himself to Steve Jobs, arguing that removing a singular creative vision from a company like Apple leaves excellent execution but no more breakthrough products, implying his own continued control is what prevents his companies from becoming merely competent rather than transformative.

The conversation shifts to the geopolitical leverage Musk wields personally through Starlink, which the interviewer credits with saving Ukraine from early collapse and which Musk notes the Ukrainian government has publicly thanked him for. He explains the recent decision to create a whitelist of approved terminals after Russian forces began using smuggled Ukrainian-ordered units in occupied territory, acknowledging this has the side effect of cutting off ordinary civilians who relied on Starlink for work or education. Pressed on the discomfort of a single individual holding power to shape a war's battlefield communications, Musk downplays his influence over the war's ultimate outcome while arguing for a negotiated peace, saying the front line has barely moved in two or three years and that diplomats calling for full Russian withdrawal are engaging in unrealistic, even callous, wishful thinking while soldiers on both sides keep dying.

He extends similar unsentimental pragmatism to his domestic political turn, admitting he 'got a little too involved in politics' and 'got carried away' with DOGE, whose goal was attacking a deficit so large that interest payments alone exceed combined spending on the Department of War and intelligence agencies. When challenged on whether abrupt aid cuts caused deaths in Africa, he flatly denies it, arguing organizations like the Gates Foundation and Mackenzie Scott's philanthropy, with tens of billions in reserves, were equally responsible for failing to fill any gap quickly.

The interview's most contentious stretch centers on Musk's social media posts about Europe, including his claim that 'civil war in Britain is inevitable' and his promotion of the film 'Citizen Vigilante,' which the interviewer, who lives in London, describes as glorifying violence against a Muslim immigrant family and painting a dystopian, inaccurate picture of the country. Musk insists he's defending 'normal people' and ordinary principles, secure borders, safe cities, sensible spending, rejecting the label 'far right' as an insult to history, while the interviewer counters with crime statistics showing London is safer than most American cities and pushes back on Musk's claim to understand a country he says he last visited years ago. The exchange grows testy, with Musk accusing the interviewer and mainstream media of being 'asleep at the switch' and denying reality, while she accuses him of exaggerating Europe's problems and lending his 250 million followers' worth of influence to parties with, in her words, 'real neo-Nazis.'

Both ultimately find narrow agreement that uncontrolled migration paired with generous welfare states creates a genuine economic pull factor, and that speech-related prosecutions in the UK are troubling, but the argument closes unresolved, with Musk asserting his 'batting average' for predicting the future is excellent and that history will vindicate him, and the interviewer maintaining he has badly overstated Europe's dysfunction.