Enterprises are confident that AI will boost efficiency, improving operations and workflows in the process. Project selection and prioritization are evolving as technology leaders find high-value use cases and try to scale projects that produce results.
Businesses currently have an average of 21 AI projects in production, according to a recent Rackspace Technology report. By the end of this year, IT leaders expect their lists only to grow.
CIOs are still determining how many AI projects are too many or too few. There isn’t, however, a magic number that enterprises should aim for. Instead, analysts said technology leaders should consider the goals, budget and overall readiness of an enterprise, its technology stack and workforce to determine an appropriate assortment.
Based on how companies are self-reporting, enterprises are landing in a wide range. Charles Schwab executives said the financial firm is investing in use cases that align with productivity and customer experience goals.
“Today, we have 40 AI use cases in various stages of development, including in use,” CEO Rick Wurster said during Charles Schwab’s summer business update last month. “Over the longer term, we believe AI will meaningfully enhance the way we serve our clients and allow us to reach our clients in an even more personalized way.”
Some organizations see value in exploring ample applications of the technology. United Airlines CEO Scott Kirby touted the company’s portfolio of projects, saying it is “probably doing more AI than anyone,” during an investor conference earlier this summer.
Business leaders see a healthy pipeline of AI projects as indicative of progress and innovation. CIOs and analysts have characterized heavy experimentation as providing necessary experience and credited the approach with better results down the line.
“I don’t know the exact count of how many are in production … but we have over 100 for sure,” Kathy Kay, EVP and CIO at Principal Financial Group, told CIO Dive. Kay said the financial services company has explored or applied the technology to contact center analytics, code tests and quality processes, documentation and customer support, among other use cases.
Enterprises often have more ideas for use cases than they can afford to pursue, given the costs and resources available.
“As part of our governance, we keep track of all the use cases and we put together what we believe the outcome will be … like cost savings or some sort of growth or elimination of a type of work,” Kay said. Monitoring keeps projects on track and allows leaders to cut the cord on dead-end use cases before costs and pressure pile up.
Widescale exploration can also be a sign that companies are early in their adoption process, according to Greg Macatee, senior analyst at S&P Global Market Intelligence 451 Research.
“Those that are a bit farther ahead … are starting to consolidate AI projects baked on what has the biggest ROI for them or what has the highest chance of success at the same time,” Macatee said. “The other side [is] we definitely talk to a lot of CIOs that have AI [fear of missing out] or are a little bit farther behind … and they’re still trying to figure out what works for them.”
Streamlining, consolidating, expanding
CIOs should be mindful of resource overextension or project bloat.
“If enterprises are getting to AI sprawl territory, then they need to step back a bit,” said Eden Zoller, chief analyst of applied AI at Omdia. “There’s not much point of treating AI like spaghetti and throwing it at different use cases, hoping it’ll stick.”
AI isn’t the right solution for every problem, and technology leaders need to consider whether investment is justified, what results they want to achieve and how ready is the organization, according to Zoller.
“Enterprises need to be far more critical and take a far more robust approach to assessment, running proof of concept before they even think about moving to full-scale deployment,” Zoller said.
As AI hype has subsided, the consolidation of projects — or streamlining — to some degree has emerged as a persistent trend. Most CIOs can’t, and analysts say shouldn’t, pursue every use case.
“People say costs are going down, but when you’re the one writing the check, it’s still pretty big,” Macatee said. When it’s time to prioritize, technology leaders should remember to align project pursuits with broader business goals.
“Know your business, measure your results and be able to communicate them,” Macatee said. “From there, you apply those results, streamline them, consolidate to what’s working well and get rid of the stuff that’s not.”
Trimming the fat has become more common as companies adopt cost-conscious approaches during the ongoing market volatility. The share of companies abandoning the majority of their AI initiatives jumped to 42% this year, up from 17% in 2024, according to analysis from S&P Global Market Intelligence published in March.
Goldman Sachs tweaked its AI investment plan as its strategy progressed.
“We started with an enormous number of [AI] use cases, and we whittled it down to the use cases that we want to spend money on,” COO and President John Waldron said during an investor conference in May.
Businesses are also leaning on project consolidation as a way to recenter and refocus efforts.
Walmart trimmed its lineup of agentic offerings to four “super agents” in an effort to streamline adoption, the company said in July.
“Once we saw how quickly teams were adopting these agents and how helpful they were, we realized agents weren’t just useful, they were essential,” Suresh Kumar, global chief technology officer and chief development officer at Walmart, said in a LinkedIn post. “But we also recognized that multiple agents — even if each one is useful — can quickly become overwhelming and confusing.”
Disclosure: Informa, which owns a controlling stake in Informa TechTarget, the publisher behind CIO Dive, is also invested in Omdia. Informa has no influence over CIO Dive’s coverage.
