As AI use expands, manufacturers continue to encounter obstacles that can slow progress.
Key takeaways
Despite these challenges, manufacturers are already using a wide range of AI capabilities.
Survey data shows manufacturers are supporting AI initiatives with meaningful investment.
New findings from an RSM survey show that while manufacturers are embracing artificial intelligence, they are doing so guided by operational realities, data readiness and return on investment.
Among the 129 manufacturing industry respondents to the RSM Middle Market AI Survey 2026, 88% said AI is already at least partially integrated into their organizations, with 32% reporting full integration across core operations and processes. Another 56% described AI as partially integrated, suggesting that adoption is broad, even if expansion is happening in phases rather than all at once.
That pattern reflects the complexity of manufacturing environments, where AI must operate across production systems and enterprise platforms that were often built long before AI was part of the conversation.
Adoption challenges rooted in manufacturing realities
As AI use expands, manufacturers continue to encounter obstacles that can slow progress. Survey respondents most frequently pointed to the following factors as limiting broader AI deployment:
37%
Security and privacy concerns
32%
Data quality, availability and lineage issues
27%
Integration with legacy systems
24%
Talent and skills gaps
These challenges are not isolated. In many manufacturing organizations, operational data is distributed across numerous systems, often with inconsistent standards and ownership. When AI initiatives attempt to derive insights across those environments, weaknesses in data quality, integration and governance become more visible.
Security concerns are also amplified by AI’s need for connectivity. As manufacturers link production, finance and operational data to enable AI, a vulnerability in one system can affect many others.
“The manufacturing environment has so much complexity because of how many systems are involved and how much data exists,” says Ryan Farlow, an industrials senior analyst at RSM US. "AI depends on seamless data connectivity, so a failure in one system doesn’t just create a localized issue; it can ripple across the entire enterprise. Security and privacy risk isn’t isolated anymore; it’s magnified.”
Where AI is gaining traction in manufacturing today
Despite these challenges, manufacturers are already using a wide range of AI capabilities. The survey shows particularly strong uptake in areas that support operational efficiency, analysis and decision making. Among manufacturing respondents:
79%
are using generative AI
63%
are using language AI
57%
are using prediction AI
56%
are using agentic AI
Looking ahead, adoption is expected to deepen rather than stall. More than 90% of manufacturers said they currently use generative, predictive and language AI or plan to use it within the next 18 months, indicating that many organizations are moving from experimentation toward more consistent deployment.
This aligns with how manufacturers describe their overall approach to AI. Nearly half (47%) said their strategy is to apply AI where it delivers clear, near‑term value, rather than pursuing broad transformation efforts before the fundamentals are in place.
AI investment: Areas of focus
Manufacturers are supporting AI initiatives with meaningful investment. Half of respondents said their organizations plan to invest at least $1 million in AI this fiscal year, and 23% expect spending of $3 million or more.
Current AI investment in manufacturing is most often focused on the following areas, respondents said:
- AI software and embedded AI solutions (61%)
- Data platforms and infrastructure (45%)
- AI research and innovation (44%)
- Risk management, including cybersecurity and model security (30%)
Most manufacturers also expect AI spending to continue rising. Eighty-four percent anticipate increased AI investment next fiscal year, suggesting sustained momentum.
At the same time, higher AI spend can require trade‑offs. Among manufacturers expecting increased investment, respondents most often said funding may be redirected from:
- Business intelligence or analytics initiatives (50%)
- Cybersecurity investments (45%)
- Legacy application modernization (40%)
Because cybersecurity is increasingly important with AI integration, however, organizations need to be strategic about changes that might reduce cybersecurity investments. AI investment decisions should align with and support broader modernization efforts.
AI integration is critical
Manufacturers reported strong results from their AI initiatives so far. Ninety-eight percent said they are satisfied with the business value AI has delivered, including 70% who described themselves as very satisfied.
Among respondents who initiated AI pilots in the past 24 months and reported moderate or limited success, the most common reasons they cited as to why AI pilots failed include:
- Data quality issues (51%)
- Integration challenges (43%)
- Security or compliance concerns (34%)
- Unclear ROI (30%)
- Adoption resistance (30%)
Integration is an especially critical point manufacturers need to focus on to achieve success.
“AI can’t just sit on top of existing processes,” Farlow says. “If AI is not built into how work actually happens, adoption and value are going to be limited.”
Companies also need to be intentional about how they communicate with employees about AI use and how they equip teams to learn new AI tools. While 88% of manufacturers agreed they have the right in‑house AI expertise, only 22% reported their AI investments are concentrated on upskilling internal talent. At the same time, 84% agreed executive leadership is more enthusiastic about AI than employees, suggesting a communication and change management gap that could slow adoption.
“You can’t treat AI like an add‑on or expect employees to figure it out on their own,” says Farlow. “Human‑in‑the‑loop processes help employees see AI as a partner, rather than something that’s going to replace them.”
AI within manufacturers’ tax functions
To better understand AI adoption within tax functions, the survey included a set of tax-specific questions. These questions were asked only of respondents who reported that their organization currently uses AI in tax, were familiar with its application, and indicated that their organization is pursuing or planning AI-enabled tax use cases over the next 12 months.
Findings show AI is gaining traction within manufacturers’ tax functions, where organizations are applying it to improve planning, compliance and decision making; 84% of manufacturing industry respondents said their organization’s tax function already uses AI tools, either formally or informally, in its business practices.
At the same time, many organizations are still working through data standardization and system integration challenges. Among organizations pursuing AI use cases in tax, 35% of respondents described their environment as advanced or leading when it comes to the organization’s technology for supporting AI-driven insights in the tax function. Sixty-five percent said their environment is fragmented, emerging or developing in that regard.
Even so, expectations for AI’s impact are high: Nearly all respondents pursuing use cases in tax (97%) anticipate that AI will moderately (41%) or significantly (56%) change the nature of work within the tax function in the next two to three years.
When provided with a list of scenarios for how respondents expect AI to change the nature of work within their organization’s tax function in the next two to three years, the top picks were:
- Strategic tax advisors: The focus shifts from compliance execution to strategic planning and business partnership (48%).
- AI orchestrators: Tax professionals primarily manage, train and oversee AI systems (45%).
- Hybrid specialists: The combination of tax expertise and AI skills becomes standard (44%).
- Tax data scientists: Roles evolve to focus on data quality, model governance and analytics (43%).
You can’t treat AI like an add on or expect employees to figure it out on their own. Human-in-the-loop processes help employees see AI as a partner, rather than something that’s going to replace them.
Ryan Farlow, Industrials Senior Analyst, RSM US
Steps manufacturers can take to strengthen AI outcomes
Based on the survey findings, manufacturers looking to expand AI impact can focus on several practical actions:
Strengthen data foundations: Prioritize data quality, integration and governance before scaling AI initiatives.
Design AI into workflows: Focus on embedding AI into end‑to‑end processes rather than deploying stand-alone tools.
Align AI with security and risk management: Ensure governance and cybersecurity evolve alongside AI adoption.
Prepare the workforce: Invest in training and change management to help employees work effectively with AI.
Use human‑in‑the‑loop models: Maintain employee oversight and judgment as AI capabilities expand.
Foundations for the future
For many manufacturers, the next stage of progress will depend less on new technology and more on reinforcing the foundations—data, integration, governance and workforce readiness—that allow AI to move from early-stage initiatives to durable enterprise capability.
RSM contributors
From AI investment to business impact
Discover how organizations are approaching AI adoption, overcoming barriers and preparing to scale enterprise value.
