Why most U.S. manufacturers are still not leveraging AI and automation

AI News


While AI and automation seem to be the industry’s biggest trends, Brian Gerkey, Intrinsic Chief Technology Officer, recently said: amazing statistics: 80% of US manufacturing facilities have zero automation.

Despite discussions about the technology’s potential benefits, the U.S. is still far from widespread adoption, much less the level of fully automated factories seen in countries like China and Japan.

“There’s definitely a lot of interest overall, but where things get difficult is execution,” said association president Jeff Bernstein. Automation Promotion Association. The group’s survey also found that while the majority of manufacturers believe in A.I. is critical to the future, yet only a few say it is widely deployed today.

Deloitte’s 2025 Smart Manufacturing and Operations Study also showed similar results. Estimation 92% of manufacturers surveyed said they believe smart manufacturing will be a key driver of competitiveness over the next three years.

However, only about 29% of manufacturers reported already using AI or machine learning at the facility or network level, and only 24% had deployed generative AI. Looking ahead to the next two years, 41% of respondents said they plan to prioritize investments in factory automation.

Bottlenecks hindering AI adoption

Tim Gauss, principal and smart manufacturing leader at Deloitte, said manufacturers are still building the foundational capabilities needed to scale AI and automation.

According to the company’s research, nearly three in four companies plan to deploy agent AI within the next two years, yet only one in five companies reported having a model with it. State of AI in Enterprise Report.

“Many organizations still use fragmented legacy systems and data. It was not configured with the use of AI in mind,” said Jasmeet Singh, executive vice president and global head of manufacturing. infosys.

According to Singh, it often comes down to digital maturity. Manufacturers that have already modernized core systems, invested in the cloud, and built strong data foundations are accelerating their transition to AI. Singh said these companies are well-positioned to scale beyond the pilot because their data is ready to support advanced use cases.

Singh added that a key change that many manufacturers are struggling with is moving from pilot projects to measurable business results. This is because manufacturers want a clear return on investment before spending large sums of money.

“In many cases, past proof-of-concept efforts have not had enterprise-wide impact, which has slowed widespread adoption,” he said.

Disappointments often stem from the way AI is implemented rather than the technology itself, which Singh refers to as using “AI for AI’s sake.”

But even these “failed” projects can be valuable, according to Bernstein, because they “lead to a clearer understanding of where the real constraints are, leading to a successful implementation the next time.”

Automation considerations

Most of today’s automation is built for highly standardized and repeatable processes, said Stefan Nasser, the company’s chief product and sales officer. essential, Google’s robotics software company. However, manufacturing often involves variation, customization, and process evolution, and automation can be expensive to test and implement.

Nassar said that even if a process were established, it might only run for a short period of time, despite the considerable bandwidth required to set it up. This may be because automation is only used for specific tasks and not extended across the factory. Companies may also realize that the technology is not viable or profitable under their current business model, or that their manufacturing needs change over time and require adaptation to new processes.

The biggest bottlenecks are often found in mid-sized manufacturers who have the resources to experiment with AI and automation solutions but tend to wait until the value is clear before making large investments. Larger organizations, on the other hand, have more resources to pursue aggressive growth, but their size and legacy systems can slow things down, Singh said.

That’s why Nusser advocates gradual adoption. He gave the example of autonomous mobile robots, which are more flexible and less risky to implement compared to traditional conveyor-based solutions that require an overhaul of the entire structure and workflow. Experts say the latter may seem nearly impossible, given that even a few days of system outage can result in millions in losses and supply chain disruptions.



Source link