Factory Robots Still Short of Autonomous Planning
A Boston Consulting Group framework places today's factory systems at Level 2 or 3, with planning-level autonomy still ahead.
Updated on Nov. 3, 2026 in Robotics
Live Poll
Do you think robots will soon change the jobs or prices you see every day?
Boston Consulting Group has published a report, "How Physical AI Is Reshaping Robotics Today -- and What Comes Next," assessing how far physical artificial intelligence has progressed on factory floors. The report describes a five-level framework for capability in industrial robotic systems and places today's factories at Level 2 or possibly Level 3. This is a research assessment, not a shipped product.
Why it matters
The framework gives manufacturers and researchers a shared yardstick for how autonomous industrial robots actually are, separating fixed-program automation from systems that perceive, manipulate and plan. Its finding that current factories sit below the planning level suggests the next gains depend on causal reasoning that BCG says no existing system has.
BCG's framework ranks physical AI in five levels, from deterministic repetitive tasks at Level 1 to workflow planning at Level 4, with factories currently at Level 2 or possibly Level 3. The report offers no benchmark figures, so the placement rests on the framework's qualitative criteria rather than measured performance.
The players
Boston Consulting Group
A global management consulting firm that publishes industry research on automation, AI adoption and manufacturing strategy.
Tilman Buchner
A partner and director at Boston Consulting Group and co-author of the report on physical AI in robotics.
The details
Level 1 systems carry out repetitive tasks through deterministic behavior, meaning they follow fixed instructions. Level 2 adds visual perception: robots fitted with cameras can identify the position and orientation of any object in three-dimensional space. Level 3 brings dexterous manipulation, letting robots mirror the function of the human hand, while Level 4 uses large language models, AI systems trained on large text corpora, to build internal models of the steps a task requires.
Timeline
BCG's physical AI report was discussed in a video dated November 3, 2026.
The Tech Race
BCG's framework gives the field a common scale for comparing industrial robot autonomy, with today's factories at Level 2 or possibly Level 3. Progress toward Level 4 planning depends on large language models, and the report's note that no system reasons like a human leaves Level 5 as the open frontier.
Manufacturers evaluating robot purchases can use the five-level framework to ask vendors which capability tier a system actually reaches. BCG does not give a availability date or pricing for Level 3 or Level 4 systems, so the practical timeline for factory deployment is not stated.
The takeaway
The report's central claim is that factories must move from perception and manipulation toward causal reasoning before robots can operate autonomously in live settings. Watch for BCG or other firms to publish measured benchmarks that test whether any system reaches the Level 4 planning tier.
Further reading
For broader context, explore the Robotics coverage on how autonomous machines are moving from fixed automation toward adaptive systems.
Source note: This article includes information reported by Supplychainbrain.
Live Poll
Do you think robots will soon change the jobs or prices you see every day?
