Artificial intelligence—Ready or not?

AI prerequisite

Takeaway

Data: Is the company’s data collected, clean and accessible to fuel AI? According to the RSM Middle Market AI Survey 2025, data quality is the top concern for AI adopters. So, it’s important to paint a clear data picture sooner rather than later. Consider asking, “How often do you encounter missing or incorrect data in your systems?” Or, “How easy is it to combine data from different departments?”

If the answers point to data silos, a lack of data governance and/or a lack of cataloging discipline, there’s a good chance that data best practices are nonexistent or immature.

That doesn’t mean the company or future AI investments are a lost cause. It just means the value creation team will have to spend time post-close centralizing data and establishing governance before achieving any meaningful results from AI. These considerations will directly affect costs, returns and value creation timelines, so investors should be aware up front.

Technology infrastructure: Are the computing platforms, integrations and tools in place to develop and deploy AI solutions at scale? If the data checks out, the technology infrastructure and integrations are up next. Modern AI workloads and solutions typically run in the cloud and must be supported by application program interfaces. So, outdated infrastructure will throw up roadblocks, and deal teams should anticipate investing additional resources before AI can be successfully implemented.

During due diligence, aim to learn what computing resources the company currently uses for AI workloads and determine the feasibility of scaling quickly. If the infrastructure is largely based on legacy, on-premises systems, manual data pipelines or ad hoc integrations, expect to spend time and resources on integrating existing systems and building an enterprise AI environment as a priority.

Skilled talent: Does the organization have the right expertise to support AI projects? Are employees being upskilled? Regardless of how good the tech stack may be, human beings with the capabilities to design, build, adopt and manage AI are still essential. Look specifically for AI or analytics expertise and ask what specific roles the company staffs. Is AI training offered? To whom and how often?

Companies that depend heavily on external resources or that do little to upskill their own people in AI likely have considerable skills gaps that will hinder success. New operating partners may want to think about how to formalize an AI training program and decide which AI-specific skills need to be hired into the company first.

Governance and security guardrails: Are governance, security and compliance guardrails in place? The more AI comes into play, the more risk vectors are introduced, from accidentally exposing proprietary data to embedding hidden biases into models. That’s why guardrails matter. Probe into whether a target has any documented AI policies or guidelines. Do they: set boundaries on acceptable AI use? Require human review for high-stakes decisions? Define practices for data privacy and model bias?

Vagueness or inconsistency on these topics probably mean you’re looking at an AI program without a strong ethical or security foundation, which needs to be addressed as a first-wave priority after closing. Drafting and rolling out a usage policy, defining AI oversight structures, and setting security and compliance standards will help protect the organization so AI can generate value instead of inviting new challenges.

Cultural openness: Is the culture ready for AI-driven change? Even the most advanced AI models and platforms won’t deliver value if the people expected to use them are skeptical, resistant or fearful. Due diligence can be a great time to explore how employees really feel about AI. Ask questions like: “What concerns have surfaced—job security, decision accuracy or broader disruption?” Or, “How has leadership communicated AI’s purpose and benefits?” Ideally, you want to get a sense that employees are curious or at least open to integrating AI into workflows, even if full-fledged buy-in doesn’t yet exist.

If instead your queries are met with apprehension, lack of clarity or low trust, cultural resistance is likely to delay or derail AI efforts. This points to an opportunity to lean into structured change management during the early days of value creation. That might include executive storytelling that positions AI as a partner rather than a threat, training to build confidence and hands-on support to show employees how AI tools make their work easier.

Aligned use cases: Is there a clear AI strategy with AI use cases aligned with business goals? The real test of AI maturity often comes down to how thoughtfully AI initiatives connect to what the business wants to achieve.

This is where industry context matters. Some sectors, like health care or financial services, are brimming with automation opportunities that can reduce manual effort and free people for more strategic work, cutting direct costs and generating additional value. Other industries may benefit more from market-facing AI tools that drive top-line opportunities, such as identifying request-for-proposal opportunities that are most likely to result in a win.