A new TCS study finds that most manufacturers see strong potential in Physical AI, but few have built the technical and organizational foundations needed to scale it beyond isolated pilots.
Robots that catch defects before they spread down the line, vehicles that detect a worker in their path and change course: Physical AI promises to move artificial intelligence off the screen and into the physical world of the factory floor. According to a new report from Tata Consultancy Services (TCS), produced in partnership with Google Cloud, the manufacturing sector is only at the beginning of that journey.
Article Body
The “TCS Physical AI Readiness Report 2026” draws on a survey of 300 senior industry leaders across automotive, aerospace and defense, electronics and high-tech, industrial equipment, and process manufacturing in North America and Europe. Its central finding: 68% of manufacturers are either not deploying Physical AI at all or remain in experimental phases. Only 32% have progressed beyond experimentation into pilots or active deployment.
The gap becomes even more apparent when looking at scaling outcomes. Just 9% of respondents report successfully scaling pilot projects into operational deployments. Another 24% describe their pilots as technically successful, but say scaling efforts are still underway. According to the company, the core obstacle is less about the technology itself and more about integrating it into existing operations.
Labor shortages and safety as leading drivers
Surveyed executives cite mitigating labor shortages (60%) and improving worker safety in hazardous environments (55%) as the leading opportunities Physical AI addresses. Enhancing quality consistency and defect detection ranks third at 49%. Expectations are particularly high for warehouse operations, where 77% of manufacturers anticipate significant or transformational impact, followed by assembly and manufacturing operations at 75%.
Among manufacturers with active deployments, AI-enabled autonomous mobile robots (AMR) and automated guided vehicles (AGV) lead adoption at 60%, tied with AI-enabled robotic arms with vision or adaptive control, also at 60%. Humanoid robots and quadrupeds, by contrast, remain at the experimental margins, with adoption rates of just 2% and 4% respectively. Broader use of these systems, the report notes, will depend on continued progress in reliability, dexterity, safety, and economics.
Legacy systems the biggest brake on scale
Asked to identify their single biggest barrier to scaling, 32% of respondents point to legacy system integration, followed by workforce skills gaps (25%) and inadequate data infrastructure (19%). The study also finds that only 10% of manufacturers have achieved highly integrated manufacturing data platforms, while 69% operate in limited, moderate, or fragmented data environments.
Governance gaps compound the challenge: 44% of respondents say no formal accountability structure exists for Physical AI failures. On regulatory readiness, 40% acknowledge they are not prepared for emerging compliance requirements, while only 12% consider themselves fully prepared.
Investment holds steady, returns remain distant
Despite these structural hurdles, investment appetite remains intact. No surveyed company plans to reduce Physical AI spending, while 26% plan to increase it. At the same time, ROI expectations stretch out over the long term: 68% expect it will take at least three years to see measurable financial returns, and 46% expect it will take more than five.
“Physical AI is a continuous loop of perception, reasoning, decision, and action,” Sreenivasa Chakravarti, VP & Global Head of Industrial Autonomy & Engineering at TCS, is quoted as saying in the report. Organizations that embed this loop across machines, systems, and people, he argues, are turning Physical AI into genuine operational advantage.
The report’s authors conclude that scaling Physical AI is fundamentally an organizational transformation challenge rather than a purely technological one. Manufacturers that align technology investment with structural readiness — in data, governance, and workforce capability — will be best positioned to lead as the market matures.
Dr. Jakob Jung is Editor-in-Chief of Security Storage and Channel Germany. He has been working in IT journalism for more than 20 years. His career includes Computer Reseller News, Heise Resale, Informationweek, Techtarget (storage and data center) and ChannelBiz. He also freelances for numerous IT publications, including Computerwoche, Channelpartner, IT-Business, Storage-Insider and ZDnet. His main topics are channel, storage, security, data center, ERP and CRM.
Contact via Mail: jakob.jung@security-storage-und-channel-germany.de