AI in the enterprise has moved past the experimentation phase. Last year, it may have been enough for business leaders to deploy AI pilots en masse. But as the market has firmed up, expectations have risen. It’s no longer enough just to use AI. Business leaders are expected to report back on progress while driving meaningful returns for their organization.
The challenge is that many are still struggling to understand where AI actually fits in their operating model. An MIT study in mid-2025 found that the vast majority of enterprise AI initiatives were failing to deliver on expectations. Even well into 2026, Gartner reports that fewer than a third of infrastructure and operations AI use cases fully succeed.
This gap isn’t about capability. It’s about clarity. Organizations aren’t struggling to access AI. They’re struggling to delineate between where it can be used from where it should (or shouldn’t) be used. That gap is translating into more and more missed opportunities for business and unrealized innovation potential.
The companies that get this balance right, understanding where and how AI should be used in partnership with human expertise and oversight, will be best positioned to excel in the years ahead. Not the ones investing in the most AI tools, but the ones making deliberate decisions about where AI operates and where humans should remain in control. That distinction will define the difference between stalled pilots and AI approaches that drive real, innovative progress and tangible ROI.
Where humans still outperform machines
AI may be changing how work gets done, yes, but humans still need to decide what matters, what’s acceptable, and who is accountable when things go wrong. Otherwise, accountability gets lost, trust erodes, and security risks run rampant (especially frightening at a time when AI-related risks continue to rise in sophistication and severity).
The most effective organizations are drawing a line between tasks that can benefit from being automated and decision-making that requires thoughtful human ownership. Even as AI capabilities grow, there are areas where they continue to fall short:
- Judgment and strategy: Intuition, institutional knowledge, and market awareness don’t live neatly in data sets.
- Context and situational awareness: Humans recognize nuance, intent, and can navigate the grey areas where rigid rules may break down.
- Empathy and trust: High-impact or emotionally charged moments (especially those involving customers or employees) demand a human-centered response.
- Oversight and accountability: Someone has to be charged with identifying failures, bias, or getting ahead of misinformation.
- Business–IT translation: As roles converge, value increasingly comes from blending technical expertise with business judgment.
- Leadership and governance: Defining policy, risk tolerance, and sticking to ethical boundaries remains a fundamental human responsibility.
AI can surface insights, recommend actions, and accelerate execution, but it can’t replace intuition, institutional knowledge, or accountability.
What smart organizations are doing differently
The organizations that succeed with AI aren’t using it to remove humans from the equation. They’re using it to remove friction. Just look at IT. AI is already delivering real value to a role traditionally bogged down by repetition and monotony.
Practitioners can now automate tasks like patching, ticket classification, workflow generation, discovery, and risk mapping. Tasks that, although foundational to IT operations, tend to eat up a lot of IT teams’ time and can lead to a whole host of issues if not properly addressed (things like downtime, missed patches, data leaks, security breaches, etc.).
Used well, AI applied in areas like these can free teams from the operational grind and allow them to focus their time on work they actually want to do, while growing business in ways that simply weren’t possible before.
Even in my own workflow, AI has become a powerful accelerator. I regularly use generative AI tools to pressure-test ideas, explore approaches, or draft early versions of content or workflows. It can take something that might have taken me a few hours to come up with organically and get me most of the way there in minutes.
But when it comes time to actually operationalize something, I often find myself pairing tried-and-tested AI tools with manual oversight. AI can suggest a path, but it doesn’t always account for the realities of how systems are wired together or what’s required to make something work reliably in production.
That’s where human ownership comes in. AI can accelerate the process, but it still takes judgment, context, and accountability to turn a good idea into something that actually works.
Humans must remain central to organizations’ AI approach
The majority of IT work will remain human-led (though AI-augmented) through the end of the decade. According to Gartner, 75% of IT work is expected to be done by humans augmented by AI by 2030 and only 25% of IT work is expected to be done by AI autonomously by that same time.
AI represents a revolutionary means of helping IT teams get more work done with less, but that doesn’t mean it should be used to replace critical business processes like strategic planning and decision making.
While AI capabilities continue to expand, their limits are getting clearer every day. Real business value will continue to come from using AI to reduce repetitive work and amplify human expertise, not replace it. Organizations that are able to find this balance won’t just move faster. They’ll build stronger teams, smarter systems, and a more durable foundation for growth than technology alone won’t be able to deliver.
AI isn’t optional. It’s quickly becoming foundational to the next generation of IT operations. But success for organizations in the AI era won’t just come from adopting more tools. It will come from making better decisions about how and where those tools are used, in tandem with human oversight, judgement, and irreplaceable expertise.
