A new Anthropic study says physical-labor automation will take decades. The industry’s loudest robot boosters are betting on something much faster.
Robots can already handle most physical tasks in American jobs, but mostly in controlled settings and at a cost too high to replace human workers. A new Anthropic study suggests that could stay true for decades. Yahoo Finance
The study’s central finding is a slow timeline. Robot prices have fallen about 3% a year since the 1990s, and if that pace holds, it would take 40 years for robots to become cost-competitive for even 10% of US work. Yahoo Finance
A clash with the bulls
That outlook runs against the optimism of the industry’s biggest names. At CES in January 2026, Nvidia CEO Jensen Huang predicted a coming “ChatGPT moment” for robotics, spanning agents, self-driving cars and humanoid robots. Tesla CEO Elon Musk, meanwhile, is racing to establish a commercial foothold for embodied AI with the Optimus robot. He has said it could go on sale as early as next year and envisions a robot in every household. Musk has also predicted at least 1 billion humanoids in use by about 2036. Nvidia and Musk’s robotic big bets will test Anthropic’s latest projections +2
Wall Street has been receptive. Morgan Stanley forecast in May 2025 that the humanoid robot market could reach $5 trillion by 2050. Yahoo Finance
Even the fast scenario is slow
The study does not ignore the bullish case. In its fast scenario, costs fall up to four times faster than the historical rate and robots learn new tasks twice as quickly. Even then, robots would not be cheaper than people for half of today’s physical work until 2050. Yahoo Finance
Other economists have reached similar conclusions. Apollo chief economist Torsten SlΓΈk wrote that widespread physical labor displacement will take decades even under aggressive projections. Yahoo Finance
What it means for Nvidia
For Nvidia, the robotics story is real but still small next to its core business. The company reported $96.2 billion in revenue for the second quarter of fiscal 2027, including $89 billion from data centers, and it does not report robotics revenue separately. One analysis calculates that even 10 million robots with $1,000 of Nvidia content each would add only about $10 billion, roughly 2.6% of annualized quarterly sales. indmoneyindmoney
Nvidia is hedging across the field. It has backed robotics companies including Figure AI, whose Series C valued it at $39 billion, and Wayve. mexc
An awkward set of relationships
The study lands amid close financial ties between the companies involved. Nvidia said in November 2025 that it would invest up to $10 billion in Anthropic, and Anthropic committed to buy $30 billion of Microsoft Azure capacity powered by Nvidia chips. Reuters reported on September 11 that Nvidia is in talks to anchor Anthropic’s initial public offering with up to $10 billion. Bloomberg added that Anthropic is seeking to raise as much as $100 billion at a valuation near $2 trillion. That valuation is a negotiating target, not a settled number. Nvidiaβs Anthropic Bet Reveals AI Boomβs Bigger Investment Qestion +2
The takeaway
The study is a forecast, not a verdict. It depends heavily on whether robot prices follow their historical decline or break from it, which is exactly what Huang, Musk and their investors are betting on. If humanoid production scales quickly, the study’s cautious timeline will look too conservative. If it doesn’t, the market’s more ambitious projections will face a hard reckoning.
