Making AI mainstream won’t be easy. Several challenges require forethought and effort from manufacturers and employees, including:
Workforce Readiness
While business leaders recognize AI’s potential, not all companies have the resources or in-house skills to deploy it. Expertise in AI is scarce, and salaries for data science and AI talent range from $100,000 to more than $300,000, according to the World Economic Forum.
“Workforce readiness is the biggest barrier to AI reaching its full potential for manufacturing,” said Courtney Steele, marketing and communications director at DeepHow. “This is relevant to frontline workers all the way to the decision-makers and executives. Each of those folks are going to struggle with all the changes affecting workforce readiness today—culture, generation gaps, and the general public understanding of AI.”
AI also raises questions about education and training. “The next generation will still need to be savvy enough and have the acumen to critically assess the information they are looking at,” Steele said. “AI can’t do that for us. We need the human factor. Education is still relevant. There are no shortcuts.”
Zayernouri agreed, emphasizing that making AI successful requires collaboration between human and artificial intelligence. “We need highly trained engineers who understand the underlying physical processes and the economic mathematical way to incorporate such data into AI platforms,” he said.
Data Quality and Management
AI systems rely heavily on high-quality, consistent data. Many facilities struggle with data collection standardization, storage infrastructure, and real-time processing. Maintaining data integrity while handling massive volumes of information from multiple sources can overwhelm traditional IT systems, requiring significant upgrades to data management infrastructure.
Cost and Infrastructure Barriers
The cost of AI remains a significant hurdle, particularly for small and mid-sized manufacturers. “The cost for a small to medium-sized company is a large barrier to entry,” Tighe said.
Brodbeck raised concerns about subscription-based AI models, asking: “Is it worth it to pay to put your data into these platforms so the software gets smarter?”
Additionally, AI’s growing energy demands present a challenge. “There will be huge competition in securing power for AI,” Brodbeck said. “The other barrier is people. Younger generations can move in and implement AI easier.”
Security and Privacy Concerns
As manufacturing becomes more connected and data-driven, cybersecurity risks increase exponentially. AI systems controlling critical production processes are potential targets for cyberattacks. Protecting intellectual property, maintaining data privacy, and ensuring operational security while keeping systems accessible presents a complex challenge for manufacturers implementing AI solutions.
Cody Schaub, economic development director at the U.S. House of Representatives, emphasized the importance of establishing AI standards for wider adoption. “From a federal perspective, we as a country need to determine how AI will affect our national security,” he said.