Robotics Breakthrough: Cost Parity Arrives by 2028

Humanoid robots will reach cost parity with manufacturing labor in developed countries by 2028. This is the real robotics breakthrough: a fact in the making, readable in the hardware cost trajectory of the last five years.

The consensus places that moment in 2035. The gap between the two dates measures seven years. The cost curve says who is right.

This represents a regime change, not a passing trend. Anyone reading robotics as an urban delivery gadget is looking at the wrong data.

The signal the market misread

On August 11, 2026, Uber sold its entire stake in Serve Robotics, the autonomous delivery robot company born inside Postmates. The move caught Serve itself by surprise.

Uber had acquired Postmates in 2020 for $2.65 billion, and Serve germinated from that robotics division, as TechCrunch documents. The tie had lasted over five years.

From the first quarter of 2022 to the first quarter of 2026, delivery volume via Uber grew for 17 consecutive quarters. In the second quarter that trend reversed, due to robot utilization below expectations. In the same period, Serve saw deliveries with another partner grow nearly 50% in a single quarter.

The deal with Uber expires in early 2027. The market read this exit as a robotics debacle. Wrong frame: Uber is abandoning a specific operating model, not the underlying technology. The distinction matters. An urban delivery model has its own constraints: route density, curb regulation, remote supervision costs. None of these touch the hardware cost trajectory. Confusing the failure of a use case with the limit of the technology is the error that generates the wrong forecasts on timing.

The cost curve says only one thing

To call something a trajectory you need at least three historical points. Solar offers them in abundance.

According to the IEA, the cost of photovoltaic modules fell by roughly 90% between 2010 and 2020. Humanoid hardware follows the same industrial pattern. Figure AI, 1X Technologies, and Apptronik show production costs declining along a parallel curve.

The causal mechanism is explicit: growing volumes, production learning, component standardization. Actuators, sensors, and compute units share the economies of scale of electronics manufacturing.

Every doubling of volumes compresses unit cost. This is the same physics that made solar cheaper than coal in a decade. Applied to humanoids, it compresses the parity horizon toward 2028.

A clarification on the limits of this comparison. Solar is a mass-market product with already-mature demand. Humanoids still have to build that volume. That is precisely the point: the curve does not start from the top of volumes, but from the bottom. Costs fall faster at the beginning of the scale, not at the end. Anyone projecting the humanoid trajectory with the decline rates of a mature market underestimates the speed of the early phase.

The cliff event: adoption that jumps, not grows

Technology adoption rarely follows a straight line. It jumps when cost crosses an economic threshold.

For humanoids that threshold is the equivalent hourly cost of manufacturing labor. When a robot costs less than an operator on a three-year amortization, demand explodes. At that point capital stops hesitating.

Cliff event: humanoid cost parity, 2028, accelerated reconversion of assembly lines. 90% of analysts are right about the present. They are wrong about the pace of change.

Three sectors that will change shape by 2030

Three categories will take a form different from the current one:

• Warehouse logistics
• Precision manufacturing
• Care and services

Warehouse logistics will fall first. Margins are thin, tasks are repetitive, return on investment is immediate. A humanoid at the parity threshold replaces seasonal labor negotiation.

Precision manufacturing will follow closely. Automakers are already testing humanoids on the lines, and cost parity will turn pilots into operational fleets.

The third front is care and services. Here cultural resistance remains high, and adoption will arrive later, toward 2031. The direction, however, remains identical. The distance between the three sectors is not random. It follows error tolerance. A warehouse accepts a measurable error rate and absorbs it in costs. Personal care does not. The higher the cost of an error, the later adoption arrives, at equal hardware cost.

My position, and what would dismantle it

My position is clear-cut: humanoid cost parity arrives by 2028, seven years before the consensus.

This thesis rests on a mechanism, not on enthusiasm. The hardware cost curve dictates the pace, and the solar trajectory provides the empirical precedent. This is a forecast about technology, with high confidence, distinct from market timing.

What would change my mind? A verifiable slowdown of the actuator cost curve, or a component supply bottleneck that freezes hardware prices for two consecutive years. In that case the date would slip toward 2031.

What it means for those deciding now

For the CTO: reassess your automation stack before parity becomes obvious. Fixed-automation contracts signed today risk rapid obsolescence.

For venture capital: the humanoid bet looks impossible at current prices, yet the cost data make it defensible. The wrong deal is the one postponed by three years.

For the Chief Strategy Officer: any three-year plan that assumes stable manufacturing labor is planning for a dissolving world. For procurement: beware of locking rigid-automation vendors into multi-year contracts, because the technology behind them will change under your feet. More analysis on our blog.

The forecast

Forecast: by December 31, 2028, at least one humanoid manufacturer (Figure AI, 1X, or Apptronik) will announce a unit production cost equal to or below the average annual cost of a manufacturing operator in developed countries.

Confidence: Medium-High. Horizon: end of 2028. Kill signal: no manufacturer reaches that threshold by the end of 2028, with hardware prices stuck above current levels.

Inevitable, not imminent: parity is written in the cost curve. The exact moment remains the terrain of the bet, and this is mine.