gshc2020.com

It is, on its face, a question about money. It is, in practice, a question about strategy, risk appetite, organizational maturity and conviction. How a CIO answers it will shape the company's competitive position for the next five years.

The Two Failure Modes

Two common themes that lead to failure:​

Overcommitment

The temptation to declare an "AI-first" strategy, allocate hundreds of millions, restructure the org chart and announce a transformation program is real, particularly when peer companies are doing the same. But capability is moving faster than any eighteen-month plan can absorb. Large, rigid bets made today are frequently obsolete before they ship. The cost is not only financial. It is the credibility lost when ambitious programs underdeliver.

Strategic Timidity

Running endless pilots, funding a small innovation team off to the side and waiting for the technology to mature before committing real capital feels prudent. It is not. The organizations winning right now are not the ones with the cleanest pilots. They are the ones who have rewired actual workflows, retrained actual people and accepted that the learning happens in production. Caution, beyond a point, becomes its own form of risk.

The CIO's job is to find the disciplined middle. Enough investment to build real capability and real momentum, structured in a way that absorbs change rather than resists it.

A Framework For Sizing The Bet

Rather than choosing a single number, the most effective CIOs are thinking about AI investment in three distinct horizons, each with its own funding logic, time frame and tolerance for ambiguity.

Horizon One: Operational Leverage

This is the near-term layer, where AI is applied to existing workflows to make them faster, cheaper or higher quality. Coding assistance, document automation, customer service augmentation, knowledge retrieval. The investment should be steady, the return measurable and the bar for funding low.

Horizon Two: Capability Building

This is the platform layer consisting of the data infrastructure, model governance, security architecture, integration patterns and the people who run them. The return is indirect and the case is harder to make, but skimping here is what causes horizon-one wins to plateau and horizon-three ambitions to collapse. Investment should be deliberate and protected from quarter-to-quarter pressure.

Horizon Three: Strategic Reinvention

This is the longer bet of new products, new business models and fundamentally rethought operating models. The investment here should be focused, not sprawling. One or two well-resourced initiatives owned by senior leaders, with clear hypotheses, defined kill criteria and the patience to let them develop.

Step By Step Not The Same As Slow

There is a common misreading of incrementalism that equates measured pacing with low ambition. It is not the same thing. The organizations moving step by step most effectively are also moving quickly. They have simply rejected the idea that speed requires a single dramatic bet.

Step by step done well looks like this. A new workflow is shipped every few weeks. Each one builds on the last, both technically and culturally. Lessons compound. The platform matures because real systems are running on it. People learn because they are doing the work, not attending a workshop about it.

The trap to avoid is the incremental version that is really just delay. Endless pilots, perpetual proofs-of-concept and the quiet conviction that next quarter will bring more clarity. It will not. Clarity comes from shipping.

What This Means For The CIO This Year

Translating this into actual decisions requires four commitments.

Commit to a funding posture.

Tell the organization clearly that this is what we are spending on operational leverage, this is what we are spending on capability, this is what we are spending on bigger bets and here is why. Predictability of investment is itself a strategic asset. It allows teams to plan and reduces the political friction around every individual decision.

Build the discipline to kill things.

A serious AI program will generate failed experiments. The maturity of the program is measured by how cleanly it ends them. Define kill criteria in advance. Without this, the portfolio fills with zombie projects that consume budget and oxygen.

Invest ahead of the curve on talent.

The single most important variable in whether an AI investment pays off is whether the people running it know what they are doing. This is not a place to defer. Hire, partner, retain and reskill aggressively.

Communicate the pace deliberately.

If the strategy is measured, say so, and say why. If the strategy is aggressive, the same. The worst position to be in is one where the board thinks the company is moving fast, the teams think it is moving slow and the CIO is privately uncertain which is true.

The Judgment That Defines The Role

There is no formula for the right AI investment level. It depends on industry, balance sheet, talent, competitive dynamics and leadership's appetite for genuine change. The CIO's job is not to find the formula but to exercise judgment, defend it clearly and adjust as evidence comes in.

Go big and you risk overreach. Go steady and you risk irrelevance. The work is calibrating between them with honesty about where the organization actually is, not what it wishes it were.

The CIOs who get this right won't be the ones with the largest budgets or boldest announcements. They'll be the ones whose pace matched their organization's capacity to absorb change, and whose conviction held steady when everyone around them wanted either more drama or more delay.

That calibration is the job. And it's the question worth answering before any other.​

Forbes Technology Council is an invitation-only community for world-class CIOs, CTOs and technology executives. Do I qualify?