Briefing at a glance:

  • According to a report released in July by consulting firm West Monroereport, companies hoping to realize value with AI should establish AI as a new operational layer, restructure workflows, link AI to business KPIs, and redesign job roles. The report is based on a dedicated study of more than 400 U.S. business leaders.
  • The report notes that executives reshaping operations face challenges including a lack of trust in technology, rapid changes driven by model releases outpacing planning cycles, evolving regulatory risks, and capacity constraints.
  • Erik Brown, senior partner in West Monroe's technology and experience practice and author of the report, believes IT leaders should not bet on a single platform but instead keep data and infrastructure as flexible as possible. Brown told CIO Dive: "When considering large AI vendors, lock-in is an extremely important factor."

Deep insights:

The AI market is rapidly evolving as vendors such as Anthropic and OpenAI continue to release new models andupdate existing versions. To avoid falling behind, CIOs should adjust planning cycles and technology strategies to adapt their environments to today's pace of change.

Brown said that by continuouslyevaluating model releases, executives may discover AI tools that significantly reduce costs while delivering similar output quality. According to a cost governance report from Mavvrik,unexpected AI costshave forced companies to rethink their AI implementation approaches.

Brown also noted that CIOs should seek vendors offering flexibility and openness, especially against the backdrop ofgrowing concerns over system lock-in. Data from the IBM Institute for Business Value shows that 7 out of 10 senior executives say switching their primary AI vendor would be challenging.

Brown said: "We need that flexibility, we need the ability to experiment consistently across domains."

For IT leaders, the conversation about AI has shifted from experimenting with the technology and exploring where it might disrupt their business to seeking tangiblereturn on investmentand proving its value to the C-suite.

Brown said: "In my view, this is the primary theme when we work with CIOs, CTOs, and product leaders. How do we truly measure effectiveness and ensure we are using AI efficiently."

In some cases, effective AI use may mean assigning basic tasks to lower-tier AI models while routing complex tasks to more powerful models. Brown said measuring output is equally critical—for example, whether a company can bring a marketable product to market with a leaner, more efficient team, and whether the actual product is more reliable.

Brown added that even with tools in place and metrics defined, companies will still face challenges around employee adoption. He said companies can often start with small use cases and teams, seeking credible people who are open to new ways of working and can have an impact.

Brown said: "We won't be replaced by AI. As with most disruptive technologies, the way we work will change dramatically. Some jobs will be replaced by AI, but that will enable humans to work more effectively. That said, there is also a human element, and the fear of 'if I start automating my own job, what does that mean for me.'"

He added that as AI use cases become increasingly complex, a lack ofreliable datacan also pose challenges for companies. Brown said building a solid data foundation while developing AI capabilities will help companies achieve their goals without slowing the pace of innovation.