Nick Clegg’s AI perspective warns of power concentration, not superintelligence

Nick Clegg wants everyone to stop panicking about machines taking over the world — and start worrying about who controls them. That’s the core of Nick Clegg’s AI perspective, laid out in a March 2026 interview with WIRED where the former UK Deputy Prime Minister and ex-Meta policy chief dismissed talk of superintelligence as overblown hype. His preferred description of the technology everyone’s afraid of? “Very versatile and very stupid.”

Key takeaways

• Nick Clegg calls fears of AI superintelligence “hand-wavy” and says the tech industry is “winding itself up into a lather” over hypothetical doomsday scenarios.
• He argues the real danger is AI power concentration among a handful of American and Chinese companies, not machines becoming sentient.
• Clegg sits on the board of Nscale, a UK-based, Nvidia-backed data-centre startup that filed for a New York Stock Exchange IPO in September 2026.
• He owns roughly 900,000 Nscale shares, worth about $18 million at the company’s last funding valuation of $14.6 billion — though reports suggest a windfall of up to $40 million if the IPO prices at a higher valuation.
• Clegg favors “agentic” AI systems built for practical tasks — booking flights, managing supply chains, negotiating contracts — over speculation about conscious machines.

Nick Clegg challenges AI superintelligence fears

Clegg’s central argument is simple: the industry is chasing the wrong monster. Superintelligence, in his telling, is a distraction dressed up as an existential threat. Coming from someone who spent years inside Meta, one of the companies most aggressively pushing frontier AI development, that skepticism carries weight.

Dismissal of apocalyptic AI scenarios

Speaking to WIRED, Clegg characterized superintelligence as “hand-wavy” — a term that suggests the concept is more theoretical anxiety than engineering reality. Today’s systems, he says, are impressive at narrow tasks but nowhere near the godlike, uncontrollable force some fear they’ll become.

Criticism of UK AI safety summits and sovereignty debates

That skepticism extends to policy circles closer to home. Clegg has previously criticized UK AI safety summits for fixating on apocalyptic scenarios instead of digging into practical opportunities and risks. He’s gone further on the question of AI sovereignty, calling British debates about the topic “slightly dishonest,” given how marginal the UK’s role is in training foundation models compared with the US and China.

The real concern: AI power concentration in US and Chinese firms

The actual problem, according to Nick Clegg’s AI perspective, isn’t machines outsmarting humanity — it’s a small number of firms cornering the technology that shapes economies and governments. He points to the growing dominance of a handful of massive US and Chinese corporations in AI development, a dynamic he sees as far more immediate than any doomsday scenario.

This framing matters because it shifts the policy conversation. Instead of debating hypothetical machine consciousness, Clegg is nudging regulators toward questions about market structure, competition, and who actually benefits when a few companies control the infrastructure everyone else depends on. His comments implicitly raise a harder question too: whether countries like the UK, or blocs like the EU, can carve out meaningful roles in the AI economy, or whether they’re destined to remain customers of technology built elsewhere.

Clegg’s stake in Nscale’s AI infrastructure boom

Clegg didn’t retreat from the industry after leaving Meta in early 2025 — he moved deeper into it. He joined the board of Nscale, a UK-based data-centre startup backed by Nvidia, putting him directly inside the infrastructure layer that benefits when governments favor pragmatic AI rules over restrictive oversight.

Board seat and financial stake

Clegg sits on Nscale’s board alongside former Meta executive Sheryl Sandberg. He owns more than 917,000 shares — about 0.12% of the company — which the Guardian reported could be worth roughly $18 million based on Nscale’s last private valuation of $14.6 billion. Reports have suggested the stake could climb toward $40 million if the IPO prices closer to a rumored $35 billion valuation, a figure the Financial Times cited this year. Both Clegg and Sandberg are paid an annual retainer of $60,000 for their board roles.

IPO filing, valuation, and financials

Nscale filed IPO paperwork in the United States on September 18, 2026, aiming to list on the New York Stock Exchange before year-end. The company was last valued at $14.6 billion in a $2 billion Series C round led by Aker ASA and 8090 Industries, and it’s reportedly seeking to raise up to $3 billion in the offering.

Revenue for the six months ending June 30 hit $140 million. During the same period, the company’s net losses climbed to $1 billion, which it attributes to the “substantial investments” needed for a “highly capital intensive” business. Nscale currently runs five datacentres across Europe and is building a further 12, including sites in the US.

Nscale’s chief executive, Josh Payne, has framed the company’s strategy around sovereign AI demand instead, arguing that countries want to build their own models and infrastructure rather than depend entirely on US or Chinese technology. “Nscale is well positioned to serve this demand as a trusted, sovereign-capable platform,” Payne said, adding that the company’s British domicile makes it “a natural, trusted partner” for Europe’s push toward AI independence.

Betting on agentic AI over sentient machines

Rather than dwelling on whether machines might someday become conscious, Clegg is steering the conversation toward something narrower and more commercial: agentic AI systems that can take actions and make decisions within defined limits. Think software that books flights, manages supply chains, or negotiates contracts — not something that thinks for itself, just something that gets work done.

That focus lines up neatly with where Clegg has placed his money. Practical, task-oriented AI is exactly the kind of application that infrastructure providers like Nscale are built to support, and it’s a far easier sell to regulators than speculative fears about machine consciousness. Whether that framing holds up as AI systems grow more capable is a question that will keep surfacing as companies like Nscale scale up and head toward public markets.

FAQ

What is Nick Clegg’s view on the threat of AI superintelligence?

Nick Clegg dismisses fears of AI superintelligence as exaggerated and “hand-wavy,” calling current AI systems “very versatile and very stupid.”

What does Nick Clegg identify as the main AI-related problem today?

He points to the concentration of AI power in a few large US and Chinese corporations as the real and immediate issue, rather than hypothetical superintelligence scenarios.

What role does Nick Clegg play in the AI infrastructure sector?

Clegg joined the board of UK-based AI infrastructure startup Nscale and owns roughly 900,000 shares, valued at approximately $18 million based on the company’s last private valuation.

What kind of AI development does Nick Clegg support?

He advocates focusing on “agentic” AI systems that perform practical tasks, such as booking flights or managing supply chains, rather than pursuing sentient or conscious AI.

Article produced with the assistance of artificial intelligence and reviewed by the editorial team.