Physical AI nears industrial scale as adoption remains limited

Only 5% see transformation in the present.

Physical AI is moving closer to large-scale industrial deployment, but most firms have yet to integrate it meaningfully into operations.

Deloitte said only 5% of firms report that Physical AI is transforming their organisation today, compared with 41% that expect it to do so within three years. The firm added that only 3% have extensively integrated the technology into operations, a figure forecast to rise to 18% within two years.

Industrial robotics remains the most mature testbed for Physical AI, according to the paper, as manufacturers and logistics operators use it to build the capabilities needed to scale intelligent systems across the value chain. More than 500,000 industrial robots were deployed in 2024, with annual installations forecast to reach 700,000 by 2028.

Deloitte said the main barriers to adoption are cost and resource requirements, cited by 41% of firms, followed by difficulties identifying use cases at 36%, talent and skills gaps at 33%, and technology or data availability at 31%.

It added that Physical AI deployment depends not only on the technology itself, but also on operational discipline, organisational learning, and workforce readiness.

Consumer, life sciences and healthcare are expected to post the highest adoption at 22% each, followed by technology, media and telecommunications at 18%, and energy, resources and industrials at 16%.

Deloitte said operations leaders should focus on technology maturity, operational readiness, and workforce design as they prepare for wider deployment.