Goldman Sachs projects 6.5 million humanoid robots shipped globally by 2035, and Teradyne has already surged 229% over the past year.
Tesla installs Optimus production lines as Elon Musk targets 10 million annual units, while TSMC guides 40%+ full-year revenue growth through 2030.
A voluntary AI slowdown flagged by Anthropic and OpenAI leaders remains the single risk that could stall Goldman's entire 6.5 million-unit build cycle.
Just released. Our analysts combed the entire stock market and named the ten best stocks to buy right now, and NVIDIA made the cut. Enter your email to see the other nine names and why NVDA earned its spot. The report is free. Enter your email and see the full list.
One number reframes the entire robotics conversation, and it did not come from a chipmaker or an automaker. It came from a bank.
6.5 million humanoid robot units shipped globally by 2035. That is the headline projection in Goldman Sachs's (NYSE:GS) new Physical AI research report, and it is a sharp upgrade to what the firm was modeling before. Goldman now sees the ramp arriving in phases: 75,000 units in 2026, 890,000 by 2030 (revised up from a prior estimate of 256,000), and 6.5 million by 2035, representing a $138 billion market opportunity led by logistics, warehousing, and automotive.
Read the number operationally. Goldman is modeling a decade of unit shipments large enough to require dedicated production lines, memory allocation, foundry capacity, factory-floor automation, and financing. The $138 billion market figure lands on the same balance sheets that already fund the AI data-center buildout, which is why the read-through cuts across chips, capital equipment, and manufacturing (we profiled seven suppliers powering that same buildout, from electricity to cooling to networking, in a free report you can grab here).
24/7 Wall St has helped investors make money for over two decades, and our top analysts just finished ranking the definitive Top 10 Stocks To Buy Now. Not the ten biggest companies. Not the ten everyone is arguing about. The ten best stocks to buy right now.
And NVDA made the cut. See who else did.
The report is free, and you can see why we think each stock is a top investment today.
Enter Your Email and See the Ten →
The company that produced the forecast is also a direct beneficiary of the capital cycle it is describing. Goldman reported record Q2 2026 net revenues of $20.3 billion and record EPS of $20.98, with CEO David Solomon telling investors, "We are in the middle of an AI CapEx super cycle where there are demands on financing into every single financing instrument in every region of the world and across every single industry."
The listed beneficiaries are trading like the thesis is already in motion. Teradyne (NASDAQ:TER), the semiconductor-test and industrial-robotics name, is up 229.21% over the past year and 96.4% year to date. Taiwan Semiconductor Manufacturing (NYSE:TSM) is up 69.14% over one year. NVIDIA (NASDAQ:NVDA) is up 23.5% over one year. Tesla (NASDAQ:TSLA) is down 18.74% year-to-date, and Intuitive Surgical (NASDAQ:ISRG), the closest listed proxy for scaled robotics commercialization, is down 34.82% year-to-date.
Every layer of the humanoid stack has a listed proxy that is already booking revenue against it. On the compute layer, NVIDIA posted Q2 FY27 revenue of $96.22B, up 105.85% year over year, and guided Q3 to $108.0B. CEO Jensen Huang told investors, "Amazon will also adopt our full physical AI stack, Omniverse, Cosmos, Isaac, and Jetson to power its fleet of warehouse robots." That is the exact application layer Goldman flags as the entry point for the 6.5 million-unit ramp.
On the foundry layer, TSMC guided Q3 2026 revenue between $44.6 billion and $45.8 billion and expects full-year 2026 revenue growth slightly above 40% year-over-year in U.S. dollar terms. Management said demand runs strong "from this day on all the way to probably 2029, 2030."
On the test-and-automation layer, Teradyne reported Q2 2026 revenue of $1.329B, up 103.9% year over year, with Robotics revenue of $100 million, up 33% year over year. CEO Greg Smith called out "physical AI applications" as the driver.
On the humanoid layer itself, Tesla is installing first-generation Optimus production lines at Fremont after decommissioning Model S and X lines, and Elon Musk described an aspirational target of 10 million units a year of Optimus 4 versus a million units a year of Optimus 3. And on the commercialization analog, Intuitive Surgical closed the quarter with almost 13,000 systems installed worldwide, proving that robotics scaling in a regulated, mission-critical setting can compound for a decade.
For retirement-focused investors, the Goldman forecast is a framing tool for long-term positioning. It sets a scale that lets long-term holders decide which layer of the stack they want to own: the financier writing the checks, the foundry printing the silicon, the test-equipment maker qualifying every chip, or the two companies actually building the robots. Near-term calendar items in the provided data include Teradyne's dividend date of September 25, 2026 and NVIDIA's $0.25 quarterly dividend paid October 1, 2026.
The risk sitting alongside the opportunity is a governance one: Anthropic employees have cited a 10% chance AI wipes out humankind within 10 years, CEO Dario Amodei has said developers need to slow down, and OpenAI CEO Sam Altman has floated an agreement among major AI developers to do so. A voluntary pause would slow the very build cycle Goldman is modeling. Until one arrives, the money keeps building for 6.5 million robots.
If you have cash sitting in your account right now, give this two minutes. After more than two decades of helping investors beat the market, our top analysts at 24/7 Wall St. put together a definitive report on the Top 10 Stocks To Buy Today. And NVDA was one of them.
They combed the entire market. It's not 10 ideas, not 10 stocks everyone is talking about, it's what their research points to as the 10 best stocks to buy right now, and it's free. Read more here and see which stocks made the list -->>
Contact [email protected] for any questions or corrections.