Updated on August 22, 2025

The threat of a decade-long gap in AI regulation may not have been the final outcome of the "One Big, Beautiful" agenda bill, but the debate it sparked is one of the most consequential takeaways. Congress has highlighted the uncertainty around how AI will be regulated, and for now, there’s still no single national standard.

Tucked into the early discussions of this sweeping legislation was a proposal to block state-level AI rules for the next ten years. While that language was ultimately left on the cutting-room floor, the underlying challenge remains: companies are left managing a patchwork of AI-related requirements that can vary widely from one jurisdiction to the next. For manufacturers adopting AI rapidly, this moment marks a shift from waiting for regulatory guidance to taking the lead on governance.

AI Is Everywhere, But Internal Oversight Hasn’t Kept Up

Manufacturing has quickly become one of the most AI-active industries, using tools for predictive maintenance, quality inspections, scheduling and supply chain planning. More than half (78%) of companies globally already use AI, with 48% adopting it within the past year. Fifty-five percent plan to use generative AI for security operations this year alone.

The speed is impressive. But it also raises questions about internal visibility. While many manufacturers now use some form of AI, a substantial number still experience challenges with its integration and lack of internal expertise to manage it. In many organizations, AI tools are added to workflows before anyone has mapped out who owns them, how they make decisions or how they’re monitored. With regulation off the table for now, companies that don’t proactively manage AI may find themselves reacting to issues rather than preventing them.

Where Risk Shows Up and Why It’s Manageable

AI tools bring real value but also new responsibilities. An AI-powered quality control system can reduce defects, but if the training data is incomplete or outdated, it may miss key issues. A supply chain model forecasting demand can streamline operations, but if it doesn’t account for recent disruptions, it may lead to overproduction or missed delivery windows.

In high-stakes manufacturing environments, even minor AI missteps can cause delays, customer complaints or lost revenue. Because AI tools often learn and evolve over time, risks aren’t limited to the initial deployment; they can shift with new data or system changes.

There’s also a growing cybersecurity dimension to AI implementation. Many AI tools connect to core operational systems and hold sensitive data. If not protected, they could become targets for manipulation or data theft. Strong internal protocols like access controls, logging and regular audits can help companies keep AI systems secure and reliable.

Waiting for Standards Isn’t a Strategy

With no single national standard in place and state-level approaches still developing, manufacturers now have a unique opportunity to lead. Creating internal AI governance is no longer about checking a box for compliance, but protecting operations, empowering teams and building sustainable adoption strategies.

At my organization, we established an AI governance council to support this work. We saw teams experimenting with generative tools and realized we needed more structure, not to restrict creativity, but to ensure clarity and consistency. The council includes product, IT, security and compliance leaders and helps us set policies, approve tools and monitor outcomes. It’s helped us keep pace with innovation without creating risk blind spots.

Centralized Oversight: A Practical Path Forward

Manufacturers often implement AI across multiple departments—engineering, production, logistics and compliance—which makes it easy for use cases to proliferate without centralized tracking. That’s why some companies are incorporating AI oversight, such as internal governance councils, into their broader governance, risk and compliance (GRC) programs.

With a centralized view, teams can:

• Map where AI tools are being used and who owns them
• Document how models were trained and what data they rely on
• Define approval processes for new use cases
• Monitor outputs for consistency, fairness and accuracy
• Audit tools periodically to ensure continued performance and compliance

This approach doesn’t slow teams down. It gives them a clear path to use new tools while ensuring the organization maintains visibility and control.

Don’t Assume Tools Come with Built-In Guardrails

It’s easy to assume that AI products come fully packaged with the protections needed to operate safely in regulated industries. However, many tools, especially generative ones, leave decisions about ethical use, data sourcing and result interpretation up to the user. That’s a risky assumption in manufacturing, where decisions often carry safety or compliance implications.

Having an internal policy around AI use – one that outlines how data can be used, who can access which tools and how performance is measured – ensures companies don’t have to rely on vendor defaults. Manufacturers can maintain consistency across the organization by integrating these policies into existing governance and security programs.

Governance Builds Trust and Long-Term Success

Some teams hesitate to introduce AI oversight because they worry it will stifle innovation. In practice, it actually does the opposite. When employees know what tools are approved, what guardrails are in place and how to raise concerns, they’re more confident using AI in meaningful ways. Employee confidence in ethical AI use means that leaders benefit too.

With proper oversight, leadership gains a holistic view of how AI is being used and can make smarter investments, prioritizing tools that are working well and re-evaluating those that aren’t. This type of self-regulation also helps future-proof the organization. When external regulations eventually arrive (and they will), companies with strong internal programs will already have many requirements in place.

The Opportunity: Lead, Don’t Wait

Manufacturers are no strangers to innovation. Many have already embraced AI with energy and creativity. However, as AI usage spreads, that early momentum must be paired with thoughtful oversight. The current lack of a single national standard and potential for varied state-level rules shouldn’t be seen as a pause, but rather as permission to lead. Companies that define their own AI governance standards now will be better prepared to scale its benefits, manage the risk it brings and adapt to future compliance standards.

It’s not about slowing down. It’s about building systems that let you move faster, with fewer surprises and greater confidence in your tools. That’s how manufacturing turns AI from a risk into a long-term advantage.

The law may be in flux. Internal oversight can’t be.