The AI Paradox: will generative tech create leisure or intense productivity?

KATHMANDU: In a landmark display of geopolitical realignment and technological ambition, tech titans Elon Musk and Jensen Huang unveiled a massive collaborative infrastructure project with Saudi Arabia this week, signaling the formal transition of the global economy from the industrial age of energy to the “intelligence age” of artificial intelligence.

Speaking at the U.S.-Saudi Investment Forum at the Kennedy Center—an event attended by President Donald Trump and Saudi Crown Prince Mohammed bin Salman—the CEOs of Tesla/xAI and NVIDIA outlined a future where computation replaces oil as the world’s most valuable resource, and where the physical constraints of Earth eventually force the data center industry into deep space.

The centerpiece of the summit was the announcement of a strategic partnership between Musk’s xAI, NVIDIA, and the Saudi state-backed AI venture, Humain. The consortium plans to construct a colossal 500-megawatt AI data center, a project Musk described as a critical step toward a post-scarcity civilization.

The New Oil: “AI Factories”

The agreement marks a pivotal moment for Saudi Arabia, a nation that has historically anchored the global energy market. Saudi Minister of Communications and Information Technology Abdullah Alswaha, who moderated the session, framed the deal as the evolution of the “92-year alliance” between the Kingdom and the United States—shifting from the export of hydrocarbons to the export of digital intelligence.

“We are moving from building oil refineries to building AI factories,” said Jensen Huang, CEO of NVIDIA.

Huang argued that the global technology sector is undergoing a fundamental architectural shift. For decades, computing was “retrieval-based”—systems simply fetched pre-recorded data created by humans. Today, the world is moving to “generative computing,” where software generates unique intelligence in real-time based on context.

“If it is generative, and every time is different, then you need AI factories all over the world to generate that content,” Huang explained. This shift necessitates a complete overhaul of global computing infrastructure, a transition Huang cited to dismiss concerns that the current AI boom is a financial bubble.

“Moore’s Law has run its course,” Huang said. “We are seeing the transition of trillions of dollars in general-purpose computing infrastructure to accelerated computing. What looks like an AI bubble is actually the modernization of the engine of the internet.”

The 500-Megawatt Bet
The newly announced facility, to be built in partnership with the Saudi entity Humain, will begin with a 50-megawatt phase before scaling to 500 megawatts. To put the scale in perspective, a typical large hyperscale data center consumes between 30 and 50 megawatts. The xAI-Humain project represents a concentration of compute power that challenges the limits of current power grids.

Musk, notorious for his aggressive timelines, initially joked that the project would be “500 gigawatts”—an impossible amount of power for a terrestrial site—before correcting himself. However, the slip revealed the true scale of Musk’s ambition, which he believes will eventually outgrow Earth entirely.

The Future of Tech: Off-Planet Compute
Perhaps the most radical “future of tech” forecast offered during the session was Musk’s assertion that the AI industry must inevitably leave the planet.

Musk argued that as AI models scale toward “unimaginable” intelligence, the power requirements will become unsustainable for Earth’s electrical grids. He estimated that within five years, the most cost-effective way to train and run AI models will be via solar-powered satellites in deep space.

“If you want to have something that is a million times more energy than Earth could possibly produce, you must go into space,” Musk said. “The sun receives overwhelmingly more energy than the Earth… In space, it is always sunny, you don’t need batteries, you don’t need glass or framing for solar panels, and cooling is just radiative.”

Musk predicted that by 2030, space-based inference nodes would undercut terrestrial data centers on price and efficiency, solving the two biggest bottlenecks facing the future of technology: power generation and thermal management.

The End of Labor?
While Huang and Musk aligned on the hardware necessities of the future, they offered diverging visions of the sociological impact of this technology.

Musk doubled down on his long-held view that AI and robotics will lead to a “post-labor” economy. He predicted that the mass proliferation of humanoid robots—specifically citing Tesla’s Optimus program—would capsize traditional economic models.

“Work will be optional,” Musk told the audience. “It will be like playing sports or growing vegetables in your backyard. You do it because you enjoy it, not because you have to.”

Musk went further, suggesting that in a future driven by AI productivity, money itself might become “irrelevant,” with the only true constraints on civilization being the fundamental physics of mass and energy. “AI and humanoid robots will actually eliminate poverty,” he declared, describing a future where every human owns a personal robot akin to “C3PO or R2-D2.”

Jensen Huang offered a more grounded, paradoxical counter-narrative. He argued that while AI dramatically increases productivity, it will not lead to idleness. Instead, it will induce a Jevons paradox—where increased efficiency leads to increased consumption of the resource (in this case, work and creativity).

“It is my guess that Elon will be busier as a result of AI, and I will be busier,” Huang countered. “When the things you do with great difficulty become simpler, you have more time to pursue the ideas in your backlog.”

Huang cited the field of radiology as a case study. Despite years of predictions that AI would replace radiologists, the profession is growing. AI made radiologists more productive, allowing them to see more patients and diagnose with higher accuracy, which in turn fueled demand for their services.

The Geopolitical Reality
The forum underscored the inextricable link between the future of technology and national sovereignty. The presence of state leaders and the structuring of the deal through “sovereign AI” clouds indicates that the next era of tech will not be defined by open, borderless software, but by physical infrastructure protected by national boundaries.

For the United States, the alliance secures a massive export market for its most advanced silicon (NVIDIA) and algorithms (xAI). For Saudi Arabia, it represents a successful hedge against the eventual decline of the oil age, utilizing its capital and land to become a central node in the global AI network.

As the session concluded, Alswaha thanked the two CEOs for pioneering the “Intelligence Age,” a sentiment that captured the essence of the forum: The future of technology is no longer just about code. It is about energy, physics, and industrial-scale manufacturing, moving from the oil fields of the Middle East to the silence of orbit.

“We shouldn’t take civilization for granted,” Musk warned in his closing remarks regarding the expansion into space. “We need to make sure to ensure that civilization has an upward arc.”

With a 500-megawatt factory breaking ground and plans to orbit the sun, that arc is now being drawn in silicon and steel.