These days, when I talk to CIOs about how their organizations are using AI, there’s a lot less excitement about new tools and the potential for higher productivity, and a lot more anxiety about what the return on investment has been so far. Because for most companies, that return has been less than expected, especially as cost per token is increasing.
Last year, IBM’s 2025 CEO study found that only 25% of AI initiatives had delivered the expected ROI over the last few years, and only 16% had scaled enterprise-wide. MIT’s NANDA initiative found that 95% of generative AI pilots delivered no measurable P&L (profit and loss) impact. More recently, SAP and Oxford Economics’ Value of AI Report 2026 found that 69% of businesses worldwide are “satisfied” with the ROI of their AI costs – but more than two-thirds think there’s still room for improvement.
It’s tempting to read those numbers as a verdict on AI’s limitations. But the research points somewhere else. When Boston Consulting Group surveyed 1,000 senior executives, it found roughly 70% of AI implementation challenges stemmed from people and processes, 20% from technology and 10% from the algorithms themselves.
The companies seeing the most value from AI? They were investing 10% in algorithms, 20% in technology and data, and 70% in people and processes.
As CIO, you don’t just own the tools. You lead the teams that determine whether the tools pay off.
My own experience with AI – integrating it into Per Scholas’s classrooms, systems and processes to achieve even stronger outcomes for our learners – has convinced me that there are three cultural traits separating teams that produce returns from teams that produce pilots: they are results-oriented, entrepreneurial and collegial.
Start with the result, and measure progress
Decide what you’re trying to change before choosing what to invest in. McKinsey’s 2025 State of AI survey found that while 80% of companies set efficiency as an objective for their AI initiatives, the companies seeing the most value often set growth or innovation as additional objectives. Those high performers are nearly three times as likely to report fundamentally redesigned individual workflows, and that redesign was among the strongest contributors to business impact of all the factors McKinsey tested.
Leaders are also getting more realistic about timelines. In TEKsystems’ State of Digital Transformation 2026 report, only 27% of organizations expected transformation ROI within six months, down from 42% a year earlier. That’s a good thing. It gives CIOs room to measure honestly instead of declaring early victories.
In practice, every AI project needs a named owner, a baseline and a business metric before it starts. If you can’t say what “better” looks like in numbers, you won’t be able to prove you got there. When we built ResumeBoost, a tool to help our alumni strengthen their resumes, our first measure was time. It cut the time staff spent working with alumni on their resumes by over a third and ensured alumni were aligning their skills and experiences with their target roles.
But measuring honestly means tracking more than one number, and being willing to act on what the different metrics tell you.
Make it safe to try, and to fail small
If you focus on results but don’t provide room to experiment, your teams will only invest in safe bets. AI rewards the opposite. A 2025 MIT Technology Review Insights survey found that 83% of business leaders believe “experiment-friendly cultures” make AI initiatives much more likely to succeed, yet 22% had hesitated to lead AI projects for fear of failure or criticism.
Hands-on experimentation isn’t a distraction from results. It’s how teams build the judgment to deliver them. And that entrepreneurial energy is most likely already in your organization. MIT’s researchers found a “shadow AI” economy in which employees are using personal AI tools even when official pilots stall. Their prescription included moving from central labs to empowered teams. The question for CIOs is whether that initiative happens inside your governance or around it.
ResumeBoost is a good example of failing small. The time savings were real, but the user experience for alumni didn’t hit the mark. Rather than keep it as a standalone tool, we folded its functionality into the broader AI coaching platform used by learners. The pilot did its job: it showed enough promise to warrant further investment, and it told us where to pivot.
At Per Scholas, that entrepreneurial energy comes from people with deep institutional knowledge. Ed Hernandez, who designed our AI-Native Classroom, has been part of the Per Scholas community for roughly fifteen years. He started as a learner and later became an instructor. Then, after more than a decade of leading our curriculum and instructional development team, Ed’s passion for AI and software products led us to create an AI solutions team and ask him to lead it. He knows where the real pain points are because he’s seen them from both sides of the classroom.
For the AI-Native Classroom, leadership’s job was to say yes, set clear guardrails and remove obstacles. But the biggest task for us was to weigh possibility against necessity. The AI-Native Classroom earned support not only because the ideas were compelling, but because they solved problems our instructors and learners actually had.
That’s the entrepreneurial culture I’d urge every CIO to build: authorize initiative, bound the risk, tie experiments to the mission and reward the people who try.
And the classroom is only part of the picture. A significant share of our AI work is staff-facing, built for the teams who run our programs and connect graduates with employers. The same principles apply there.
Candor helps, too. People don’t experiment freely with a technology they suspect could soon replace them. Be transparent about what the goals are, and what it would mean for your organization to achieve them.
Build collegial, cross-functional teams
AI can’t replace collaboration. What it can do is make collaboration more valuable by bridging silos.
The companies pulling ahead treat collaboration as infrastructure. Accenture’s research on “Talent Reinventors,” organizations that advance people and AI together, found they delivered revenue growth 1.8 percentage points higher and profit growth 1.4 points higher than peers in 2025.
Collegiality means pairing technologists with the people who own the workflow, rather than delivering tools to them. We built our AI-Native Classroom with our instructors, not for them. We took the same approach with Sherlock, a job-sourcing tool our AI team built with the staff who connect our graduates with employers.
Finding roles for learners used to mean opening a dozen job boards, rerunning the same searches and weeding out staffing firms’ reposts of the same opening. Sherlock now reviews more than 125,000 postings a day and delivers one feed of roles that match our training tracks and each campus’s recruiting footprint.
But it was designed around how that team already works. Saved roles go on a list shared across each campus, so one colleague’s find becomes everyone’s lead, and they export in a format built for the Salesforce job orders staff already create. Staff also tune the feed themselves by flagging roles that don’t belong and ones they know exist but never appeared. Three weeks after launch, staff at nearly three-quarters of our campuses had logged in, weekly users had more than doubled and users saved more than 100 hours they would otherwise have spent sourcing jobs.
Seven weeks in, those same staff are telling us where Sherlock falls short, and we’re now scoping out the enhancements they’ve asked for. It’s an early success, not a finished story. Staying successful requires incremental implementation driven by an open dialogue between the people who build the tools and the people who use them.
The talent you hire is the culture you get
None of this requires a new platform. It requires CIOs to treat culture as part of the AI investment. AI is a force amplifier, not a force generator: it magnifies whatever already exists in a team. With a strong culture, AI becomes an extension of what makes an organization distinctive. With a weak one, the same problems surface, only faster.
Culture also shapes who you hire and how you develop them.
The employers we work with increasingly see talent as something to build, not just buy. As Equinix’s chief people officer put it when describing the company’s investment in early-career pathways, “We’re not just competing for talent; we’re creating it.” When we prepare learners for high-growth careers, technical skills are the baseline. What employers tell us sets people apart is whether they can define a problem, try something new and work across a team to solve it.
Those are the same traits that decide whether your AI budget becomes a line item or a return.