The Robot Takeover Begins: How Elon Musk and Silicon Valley Are Racing to Make Human Labor Optional

The Pivot That Ended Two Flagship Cars

On January 28, 2026, during Tesla's Q4 2025 earnings call, Elon Musk announced that the company would end production of both the Model S and Model X in the second quarter of 2026. According to CNBC and The Verge, the reason was direct: Tesla needed the production floor space at its Fremont, California factory to build Optimus humanoid robots. The Model S had been in production since 2012; the Model X since 2015.

The cancellation of two longtime flagship vehicles in favor of robots marked the clearest signal yet that Musk's robotics ambitions have moved from side project to core business priority. Per The Verge, Tesla's earnings report stated that the third-generation Optimus robot — "meant for mass production" — would be unveiled in Q1 2026 and that a first production line would begin operating at Fremont before the end of the year. Tesla has stated a long-term capacity target of one million robots per year.

Musk has previously predicted that Optimus robots would work not only in Tesla's own factories but eventually as home assistants and in medical settings. He has also said Tesla would begin selling Optimus to external customers, though no commercial sale date has been confirmed. According to The Verge, Musk's $1 trillion pay package from Tesla shareholders requires him to build at least one million robots — aligning his personal compensation directly with the robot deployment timeline.

What "Physical AI" Actually Means

The Washington Post, in a March 27, 2026 report, described a broader Silicon Valley movement around what executives are calling "physical AI" — the application of the same large-model advances that powered ChatGPT to robots that operate in the physical world. Unlike software-based AI, physical AI must perceive 3D environments, plan sequences of movements, and execute precise manipulation tasks — all in real time, in environments that don't hold still.

The logic behind the investment wave is straightforward: software AI has transformed industries with information-based jobs, but the roughly 65 percent of the global workforce in physical labor — manufacturing, logistics, construction, agriculture, healthcare support — has been almost entirely untouched. Physical AI is the proposed solution to that gap.

Musk has framed this in the most expansive terms possible. At Tesla's shareholder meeting in November 2025, he said the company's mission was "sustainable abundance via A.I. and robotics — that's the future we're headed for," according to the New York Times. At a US-Saudi Investment Forum in Washington in January 2026, per Fortune, he predicted that "in the next 10 to 20 years, work will be optional," comparing employment to maintaining a vegetable garden — something people might do for enjoyment rather than necessity.

Where It's Actually Working: The BMW Evidence

While Musk's predictions remain predictions, Figure AI — a privately held robotics company based in Sunnyvale, California — has published hard operational data from a real factory deployment. According to Figure AI's own published report, its Figure 02 humanoid robot completed an 11-month deployment at BMW Group Plant Spartanburg in South Carolina performing sheet-metal loading on an active assembly line.

Figure AI reported the following results: more than 90,000 parts loaded; more than 1,250 hours of runtime; contribution to the production of more than 30,000 BMW X3 vehicles; an estimated 1.2 million robot steps (approximately 200 miles walked). The robots ran 10-hour shifts Monday through Friday. The task required placing parts within a 5-millimeter tolerance in under 2 seconds per placement.

BMW Group confirmed a subsequent expansion of the program to its Leipzig, Germany facility, marking the first humanoid robot deployment at a European BMW plant, per BMW Group's press releases. According to Automotive News, Figure AI's units logged more than 1,250 hours at Spartanburg, moving more than 90,000 parts over 10 months before the program expanded.

Following the Spartanburg deployment, Figure AI announced that it was retiring the Figure 02 and transitioning to the Figure 03, incorporating lessons learned about forearm hardware reliability, wrist electronics architecture, and thermal management under sustained daily operation.

Boston Dynamics and the Competitive Landscape

Tesla and Figure AI are not alone. Boston Dynamics, owned by Hyundai, unveiled the production version of its fully electric Atlas robot at the Consumer Electronics Show in Las Vegas on January 5, 2026, according to Boston Dynamics' official press release. The new Atlas is designed for industrial work, material handling, and what the company calls "intelligent automation."

The Register, reporting on the CES announcement, noted that Boston Dynamics beat Tesla to a production-ready robot reveal, a notable reversal given Musk's years of public promotion of Optimus. Tesla's robotics head Milan Kovač departed the company before the Gen 3 announcement, according to The Verge.

Other entrants include 1X Technologies (backed by OpenAI), Agility Robotics (backed by Amazon), and a set of Chinese manufacturers that Goldman Sachs Research in November 2025 described as "aggressively building humanoid robot capacity ahead of orders," with Chinese suppliers benefiting from a domestic chip and manufacturing ecosystem that enables faster iteration.

The Market Forecast

Goldman Sachs Research has projected that the global market for humanoid robots could reach $38 billion by 2035, a figure the firm's analysts noted was larger than their earlier estimates. In a February 2024 analysis that became a baseline for subsequent forecasts, Goldman Sachs estimated that humanoid robots could fill approximately 4 percent of the U.S. manufacturing labor shortage gap by 2030 and approximately 2 percent of global elderly care needs over the same period.

Goldman Sachs research published in November 2025 noted that 2026 was expected to be a breakout year for humanoid robot orders and shipments, driven partly by Tesla's Optimus Gen 3 reveal and partly by Chinese manufacturers scaling up supply chains ahead of expected demand.

As of early 2026, industry analysts and consulting firms tracking the sector reported that humanoid robot deployments were concentrated in manufacturing and automotive assembly (approximately 35 percent of deployments), logistics and warehousing (approximately 25 percent), and research settings (approximately 15 percent), according to sector analysis published by Robozaps, citing Goldman Sachs and industry deployment data.

What Hasn't Worked Yet

The New York Times, in its February 27, 2026 profile of Musk's "sustainable abundance" vision, noted that Tesla's Optimus robots "have been displayed at company events but have shown limited mobility." Previous public demonstrations included what The Verge described as "teleoperations missteps" — instances where Optimus robots at events appeared to be remotely operated rather than autonomous, raising questions about how much of the robot's performance was real-time AI versus human remote control.

Figure AI's own BMW deployment report acknowledged hardware failure points, specifically the forearm — citing tight packaging, thermal constraints, and wrist actuator complexity as areas that required a complete redesign between the Figure 02 and Figure 03 generations. The company set a target of zero human interventions per shift; while not explicitly stated whether this target was met consistently, the report framed it as an ongoing engineering challenge.

More broadly, the economics of humanoid robot deployment remain unproven at scale. A robot that costs hundreds of thousands of dollars to manufacture, requires maintenance, and operates at 84-second cycle times on a single repetitive task faces a long road before it can match the flexibility and cost-per-hour of a human worker across diverse tasks. The Goldman Sachs $38 billion projection is a market size estimate for 2035 — not evidence that the economics are favorable today.

The Labor Displacement Question

Missing from most industry coverage of the humanoid robot race is a direct accounting of what happens to workers displaced by the technology. Musk's "work optional" framing presents this as a benefit — freedom from labor. Critics point out that a post-scarcity society requires either a redistribution mechanism (such as a universal basic income) or a continuation of wage-based employment, neither of which Musk or Tesla has proposed in concrete policy terms.

The BMW deployment involved tasks previously performed by human assembly workers. Figure AI's report describes robots taking over "an associate" who had previously performed the sheet-metal loading manually. BMW's official communications have framed humanoid robots as tools to "reduce physical strain on workers by taking on higher-risk and repetitive tasks," implying a worker redeployment rather than replacement model — though neither BMW nor Figure AI has published data on what happened to the workers previously assigned to the automated stations.

The International Labour Organization has not yet issued a formal assessment specific to humanoid robotics and labor displacement. The economic modeling that exists — including Goldman Sachs' projections — treats humanoid robots primarily as a productivity and market size story, not a displacement study.