Elon Musk, Jeff Bezos, and a growing circle of billionaire tech investors are bankrolling humanoid robots designed to work factory floors, handle warehouse logistics, and eventually enter homes. The push is no longer theoretical. Robots built by well-funded startups are already running production shifts at major automakers, and Musk has set a public deadline to sell Tesla’s own humanoid units by late 2027. The speed of these deployments, and the scale of capital behind them, is forcing a harder question about what happens to the human workers these machines are built to replace.
Musk Sets a Sales Deadline for Optimus
During Tesla’s Q4 2025 earnings call, Musk told investors the company would sell humanoid robots by the end of 2027. The timeline places Tesla on a roughly two-year sprint from early internal factory use to external commercial sales. Musk has a long record of announcing ambitious deadlines and missing them, from Full Self-Driving timelines to the Cybertruck launch. But the Optimus program now has something his earlier promises lacked: working prototypes already performing tasks inside Tesla’s own facilities.
On the same call, Tesla executives outlined plans to increase spending on AI and robotics heading into 2026, framing Optimus as central to the company’s next growth phase. The earnings transcript shows Musk positioning the robot not as a research experiment but as a product line with a commercial target date. Whether Tesla can hit that date is an open question, but the financial commitment is real and public, and Musk has tied the company’s long-term valuation story to a future in which humanoid robots contribute significant revenue.
Musk has also argued that Optimus could eventually be more valuable than Tesla’s vehicle business, suggesting a world where fleets of robots are leased or sold into factories, warehouses, and perhaps homes. That vision depends on rapid progress in both hardware and AI, and it assumes regulators will not significantly slow deployment. For now, investors are being told to expect visible milestones over the next two years as Tesla moves from internal pilots to external customers.
Bezos-Backed Figure AI Already on the Factory Floor
While Musk talks timelines, Figure AI is already logging production hours. The startup, which secured backing from Jeff Bezos and partnered with OpenAI, signed a commercial agreement with BMW Manufacturing to place its general-purpose robots directly into automotive assembly. Figure frames its machines not as specialized tools but as flexible labor, capable of adapting to different tasks on the same line.
That framing matters because it signals an intent to compete with human workers across a range of roles, not just the narrow, repetitive motions that traditional industrial robots handle. Figure’s commercial agreement with BMW is structured around bringing these general-purpose robots into real production environments, not pilot labs or demonstration stages. The company has stated its goal is workplace and home deployment at scale, presenting its robots as a new category of labor rather than a niche automation add-on.
Figure’s partnership with OpenAI is intended to accelerate the robots’ ability to understand natural language instructions and adapt to new tasks. In theory, that could make it far easier for a factory manager to reassign a robot from one station to another with minimal reprogramming. In practice, it also lowers the barrier to replacing workers in roles that were previously protected by their need for on-the-fly judgment and communication.
BMW’s Pilot Produced Hard Numbers
The strongest evidence that humanoid robots can function in real manufacturing comes from BMW’s own operational data. At its Spartanburg, South Carolina plant, BMW ran a pilot program using Figure AI’s Figure 02 robots. The results were specific: the robots supported production of more than 30,000 BMW X3 units and moved more than 90,000 components during the trial. They ran 10-hour shifts, Monday through Friday, matching the cadence of human production workers.
Those numbers strip away much of the skepticism that has surrounded humanoid robotics for years. Moving 90,000 components across thousands of vehicle builds is not a tech demo. It is sustained, measurable output inside a facility that ships real cars to real customers. BMW has since announced plans to deploy humanoid robots in production in Germany for the first time, expanding beyond the U.S. pilot into its home manufacturing base and signaling confidence that the technology can scale.
For BMW, the appeal is straightforward: robots that can plug into existing workflows, use the same tools and infrastructure as people, and operate on the same schedules promise a way to increase flexibility without redesigning entire plants. For workers, the same traits raise the prospect that any task involving moving, lifting, fetching, or basic inspection could eventually be handed off to machines.
Why “General Purpose” Changes the Threat Calculus
Traditional factory robots are bolted to the floor and programmed for a single task, like welding a seam or painting a panel. A general-purpose humanoid robot, by contrast, is designed to move through a facility, pick up different objects, and switch between assignments. That flexibility is what makes the current wave of investment different from decades of prior automation.
If a robot can only weld, it threatens welders. If a robot can carry parts, load bins, inspect assemblies, and adapt to new tasks through software updates, it threatens a much wider range of production roles. Figure AI’s pitch to BMW leans heavily on this idea, describing its machines as labor that can be redeployed rather than reprogrammed. The distinction is not just marketing. It reflects a design philosophy shared across the billionaire-backed robotics sector: build machines that can do what people do, in the places people work, on the schedules people keep.
This is the gap in current public discussion. Most coverage of humanoid robots focuses on whether the technology works. The Spartanburg pilot answers that question with concrete production data. The harder question, and the one that labor advocates, regulators, and workers themselves have barely begun to address, is what happens to the 10-hour-shift workers when the robots running those same shifts get cheaper and more capable each year. Without policies on retraining, job transition, and bargaining over automation, the benefits of these gains are likely to flow primarily to investors and executives.
The Money Trail Points in One Direction
The pattern across these investments is consistent. Musk is funding Optimus through Tesla’s balance sheet. Bezos has personally backed Figure AI. Microsoft and other tech giants have joined the funding rounds that give these startups the capital to move from prototype to production. OpenAI’s partnership with Figure AI adds another layer, connecting the most prominent name in artificial intelligence to one of the most aggressive pushes in physical robotics.
None of these investors are framing their bets as experimental. The language in earnings calls, press releases, and partnership announcements consistently describes near-term commercial deployment, not long-horizon research projects. Industry-focused outlets such as PR distribution platforms are filled with announcements that emphasize production use cases, customer contracts, and deployment timelines. A login portal for investors and clients at a major release service underscores how central these communications have become to shaping market expectations.
The sums involved are large enough to matter for labor markets. Capital flowing into humanoid robotics is not chasing incremental efficiency gains at the margins; it is targeting the core cost structure of manufacturing and logistics, which is dominated by wages and benefits. When the people writing the checks are among the wealthiest individuals and most powerful corporations in the world, their shared thesis (that general-purpose robots will be economically transformative) carries outsized weight.
Labor groups and policymakers are only beginning to respond. Some unions have started to negotiate language around automation, seeking advance notice and retraining commitments when new technologies are introduced. But the pace of investment suggests that, absent stronger guardrails, decisions about where and how humanoid robots replace people will be made primarily in boardrooms and venture capital meetings.
A Future Built on Choices, Not Just Technology
The arrival of humanoid robots on factory floors is not a distant science-fiction scenario; it is already happening in Spartanburg and in Tesla’s own facilities, with timelines for broader deployment measured in a handful of years. The technology will continue to improve, but the crucial questions now are political and economic rather than purely technical.
Societies can choose to treat humanoid robots as tools that augment human workers, taking on dangerous or exhausting tasks while people move into higher-value roles. Or they can allow robots to become one-for-one replacements, shrinking payrolls and concentrating the gains from automation in the hands of a small investor class. The direction of current investment—from Musk’s Optimus program to Bezos-backed Figure AI, suggests that the default path will favor cost-cutting and scale unless countervailing forces intervene.
That makes the present moment a hinge point. As general-purpose robots move from pilots to production, there is still time to debate how their benefits and burdens are shared, and to build rules that reflect more than the priorities of the billionaires funding them. The machines are coming either way; what they mean for the people they stand beside, or eventually replace, will depend on choices made well before 2027.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
