Factories Still Lag at Robot Reasoning Levels

A Boston Consulting Group report maps five capability levels for physical AI and finds factory systems have not advanced beyond Level 2 or possibly Level 3.

Updated on Nov. 3, 2026 in Robotics

Bold flat-color editorial illustration of a single industrial robot arm above a conveyor, beside a stepped row of gray cylinders representing rising capability levels. — Boston Consulting Group's report finds factory robots still lack the causal reasoning needed for autonomy and remain stuck at Level 2 or possibly Level 3 of physical AI capability. AI Illustration. Upload story photo >

Live Poll

Do you think robots will soon change the jobs or prices you see every day?

Boston Consulting Group published a report on the progress of physical artificial intelligence on the factory floor. It describes five levels of capability and concludes that today's factory systems have not progressed beyond Level 2 or possibly Level 3.

Why it matters

The report indicates that autonomous industrial robots still lack the reasoning needed to operate independently in live settings. Reaching higher capability levels will require fundamental innovation, not incremental refinement.

BCG defines five levels, with Level 2 adding camera-based identification of object position and orientation in three-dimensional space and Level 4 using large language models to build mental models of task steps. The report does not quantify how many factories operate at each level, and it is unclear how the framework is measured in practice.

The players

Boston Consulting Group

A global management consulting firm that publishes industry research on emerging technologies and their business applications.

Tilman Buchner

A partner and director at Boston Consulting Group who co-authored the report on physical artificial intelligence.

The details

Level 1 covers deterministic, repetitive robot tasks, Level 2 adds visual perception, and Level 3 introduces dexterous manipulation that mirrors the human hand. The report states that automated systems will need causal reasoning to operate autonomously, a capability no current system is described as having.

November 3, 2026: BCG report on physical AI in factories discussed in video

The Tech Race

BCG's five-level framework gives the industry a shared yardstick for physical AI, and its finding that factories sit at Level 2 or possibly Level 3 shows the gap between current deployments and the reasoning-capable systems that competitors are pursuing. The framework marks a measurable baseline for tracking progress against rival automation efforts.

For U.S. manufacturers and workers, the report signals that robots capable of independent reasoning on the factory floor remain a future prospect rather than a current capability. Plant operators planning automation upgrades should expect current systems to handle repetitive and visually guided tasks rather than autonomous decision-making.

The takeaway

Factory automation is advancing in measurable steps, and the gap between today's visually guided robots and reasoning-capable systems is the key benchmark to watch. Organizations evaluating robotics investments should weigh current capability levels against realistic timelines for higher-level autonomy.