A whopping 3% of boardroom execs spy productivity and costs savings from AI

The hype train for AI is still chugging along, but the numbers coming out of Gartner’s latest HR Symposium are a cold splash of water on the boardroom’s rosy forecasts. Three‑quarters of corporate directors are convinced AI will slash costs and boost productivity, yet a paltry three percent of firms have actually hit those targets. Meanwhile, 38 percent of companies are already over budget on AI spend, and a staggering 85 percent plan to pour even more cash into the pot next year. As a founder who’s been wiring up data centres and keeping the lights on for a decade, I see this gap between expectation and reality as a warning sign for anyone thinking AI is a free‑ride to the bottom line.

Why the Boardroom is Over‑Optimistic

The Gartner data makes one thing clear: senior executives are wildly out of sync with the rest of the organisation on where AI actually creates value. At least 75 percent of business leaders are out of step with the broader workforce, according to the analyst. That misalignment stalls transformation, muddies workforce planning, and ultimately blocks any real return on AI projects. In my own experience, when senior management pushes a shiny new tool without a clear line‑of‑sight on the problem it solves, the result is a half‑baked pilot that never scales.

What’s more, the “great opportunity” that Gartner’s senior director CV Viverito mentions – shifting employee time into new capabilities and growth sources – is still a pipe‑dream for most firms. Only about one in five employees’ time is actually being freed up by AI, according to the symposium. That means the majority of staff are still stuck in the same routine tasks, with AI barely nudging the needle on capacity creation.

The Real Cost of Over‑Spending on AI

Over‑budget AI projects are the new norm. The 38 percent figure shows that a sizable slice of the market is already feeling the pinch. Companies are not only spending more than they planned, they are also planning to spend even more in 2027. This escalation is happening while the ROI remains elusive. A Forrester report cited in the same briefing predicts AI will make up roughly 20 percent of all business technology spend next year, yet more than half of CIOs are likely to miss their ROI targets. The math doesn’t add up unless you factor in hidden costs – training, integration, and the inevitable need for specialist talent.

From the trenches, I know that hiring scarce AI specialists is a luxury most independent hosting providers can’t afford. Gartner’s advice to spread existing AI expertise across the business rather than wait for scarce talent is sound, but it’s also a tall order when you’re already stretched thin on engineering resources. The reality is that many firms will end up paying premium rates for consultants or over‑paying junior staff to fill the gap, inflating the cost base without delivering the promised savings.

Training: The Hidden Expense No One Talks About

Viverato points out that the struggle isn’t creating capacity – it’s making that capacity usable. The next step, according to the symposium, is upskilling employees to take on higher‑value, AI‑enabled work. That sounds noble, but it translates directly into more spend on training programmes. In my own data‑centre operations, every time we introduced a new automation platform, we had to budget for weeks of intensive upskilling, and the learning curve was steep enough to cause a temporary dip in service levels.

The implication for independent providers is clear: you can’t just buy an AI tool and expect it to magically free up staff. You need a structured training pipeline, and that pipeline costs money, time, and managerial bandwidth. Companies that overlook this will find their AI projects stuck in a perpetual “pilot” phase, never moving to production where the real cost savings could be realised.

The “Magnificent Seven” vs. the Rest of the S&P 500

Private‑equity firm Apollo Global Management highlighted a stark contrast: the “Magnificent Seven” tech giants are reaping the rewards of AI spending, while the rest of the S&P 500 shows no sign of AI‑driven productivity growth. This divergence underscores a structural advantage for the mega‑players that can afford massive data lakes, proprietary models, and dedicated AI research teams. Smaller firms, especially independent hosting providers, simply don’t have that scale.

What does this mean for the rest of us? It means we need to be realistic about the value AI can deliver at our scale. Rather than chasing the same breadth of AI services that the big tech firms are offering, we should focus on narrow, high‑impact use cases that directly touch our core revenue streams – like predictive capacity planning, automated ticket triage, or anomaly detection in network traffic. Those are the areas where a modest AI investment can actually move the needle without requiring a multi‑billion‑dollar R&D budget.

HR’s Role in Taming the AI Beast

Gartner’s recommendation that HR help executives agree on the top AI bets is more than a feel‑good statement. In practice, HR can act as the bridge between strategy and execution, ensuring that AI initiatives are aligned with the organisation’s talent capabilities and that managers have the authority to redesign work around the technology. In my own operations, we set up a cross‑functional AI steering committee that included HR, engineering, and finance. This committee vetted every AI project against a set of criteria: clear business outcome, measurable KPI, and a realistic training plan.

When HR is left out of the loop, AI projects often become siloed experiments that never integrate with the broader workflow. The result is a collection of proof‑of‑concepts that sit on a server, gathering dust while the organisation continues to spend on licences and consulting fees. A disciplined, HR‑backed governance model can prevent that waste and keep AI spending in check.

Practical Steps for Independent Providers

Given the data, here’s a no‑nonsense playbook for hosting firms and other mid‑size tech outfits:

1. Identify the highest‑impact work streams. Use a simple ROI matrix to rank tasks that are repetitive, data‑rich, and have clear cost implications. Focus AI spend on those, not on flashy “innovation labs”.

2. Leverage existing talent. Upskill your current engineers and ops staff rather than hunting for scarce AI specialists. Pair them with open‑source models and internal data to build incremental value.

3. Set realistic capacity targets. Expect that only about one‑fifth of employee time will be freed up by AI in the near term. Plan your productivity gains accordingly and avoid over‑promising to the board.

4. Implement strong governance. Involve HR early to define the skill gaps, training pathways, and to align AI projects with strategic priorities. Use a steering committee to keep spend under control.

5. Measure and iterate. Track the same KPIs you used to justify the investment – cost per ticket, mean time to resolution, utilisation rates – and be prepared to kill projects that don’t hit targets within a defined horizon.

Bottom Line: AI Is Not a Silver Bullet

The Gartner numbers are a reality check: boardrooms are dreaming big, but the ground reality is that AI is still a costly, complex undertaking with modest early returns. The 3 percent success rate and the over‑budget statistics should make any founder pause before signing off on a multi‑million‑dollar AI rollout. The real opportunity lies in disciplined, targeted use of AI to free up a slice of employee capacity and to enable higher‑value work – but even that will require a concerted effort on training, governance, and realistic expectation‑setting.

For independent hosting providers and other mid‑size tech firms, the path forward is clear: stop chasing the hype, focus on narrow high‑impact use cases, and build AI capability from within. That’s the only way to turn the AI spend surge into a genuine competitive advantage without getting swallowed by the same over‑budget pitfalls that are haunting the broader market.

— Allan Ali, Founder

This article was produced with AI-assisted research and editorial support. Reporting is based on the source material cited below. Sources: The Register; theregister.com; Global1.News (28 October 2026).

By Allan Ali, Global1.News