How CIOs can overcome the AI trust gap

If you’re like most CIOs, your CEO’s viewpoint of and strategic outlook for AI has likely shifted several times over the past two-plus short years, perhaps even in contradictory ways. Initial doubts or disregard may have given way to full-out FOMO. Demands for copilots everywhere in everyone’s hands may have experienced pullback as the bills and questions of value mounted. Rampant experimentation turned to targeted use cases. Pressure for widespread AI adoption narrowing as “tokenmaxxing” and “tokenomics” entered the lexicon.

Now, with frontier labs showing what can go wrong with “agent misalignment,” AI failures hitting the news, and questions of liability still lingering, trust in AI — and their CIO’s AI strategy — is becoming a key CEO concern (once again).

As the executive setting the company’s course for AI, the CIO is expected to make every AI project pay off, to instill trust in AI for all stakeholders, and to guarantee that nothing unforeseen and unfortunate happens. As a result, the stakes are high, with CIOs already blamed for any AI misstep and, according to IDC, at risk of future firings as lawsuits, fines, and AI incidents rise through 2030.

The only way for CIOs to meet this moment is to establish governance, security, data quality, AI model quality standards, and metrics for business ROI, with a clearcut means of shutting systems down if systems become a risk.

Doing this will take time and money at a time when CEOs are being pressed to show results from AI. It will also require clear and compelling explanations of why all these steps are necessary, how these steps fit into a total corporate AI strategy, what checkpoints can be set to assure that business metrics and investment return goals are met, and how AI should be addressed in the corporate risk management plan.

Here are the essential steps CIOs should take to build trust in their companies’ AI implementations.

1. Require quality data, quality AI models, and business-relevant metrics before projects are launched

To date, many organizations have run their AI projects in “sandbox” mode. They try out different AI models and algorithms, without paying as much attention to the data and the metrics. Here, the most quoted evidence is from MIT’s 2025 AI survey, which found that enterprises had invested nearly $40 billion for generative AI, with 95% of those projects reported as yielding no return.

This year, AI’s honeymoon period has shown signs of being over. CEOs are expecting not only results but returns on their investments. This will only grow louder in 2027. And the only way to get results is to ensure that AI projects use high-quality data, that AI models are developed and maintained to the highest quality, and that clear metrics for investment return that tie to specific business use cases and outcomes are in place.

CIOs must preemptively take the lead in setting AI culture and work standards. Only from that position can trust follow.

2. Run tight, controllable projects

Recent incidents involving OpenAI, Anthropic, and Google have shown what can go wrong when AI goes rogue. No enterprise CIO wants this to happen on their watch.

To avoid this, AI projects should be structured with domains of knowledge and functions limited to the environments that they operate in (e.g., support for a call center). These AI projects should not try to do too much at first. As confidence is gained with the AI and its results, scope can be expanded.

A slow, proof-based approach to agentic AI will be essential to ensuring stakeholder trust.

3. Have an AI shut down valve

If an AI project goes awry, do you have a way to shut it down and mitigate any damage?

How to turn an AI system off should be one of the first tasks defined in any AI project plan. It’s equally important to take this one step further — by incorporating shutdown and restoration steps for every AI system in the enterprise. This can be done by including AI in corporate disaster recovery and risk management plans.

For a more granular look, see our partner publication CSO’s “Why you need a reliable AI agent kill switch” and “Securing AI agents: Key controls and best practices.”

4. Consider engaging an auditor to monitor AI security, governance, and compliance

There will always be corporate functions — sales, marketing, engineering — that will want to skip governance and get to running AI systems. In companies that stress R&D, there might even be some resistance from the board and CEO, who want to avoid “red tape” and slowdowns.

Regardless of company culture, it will be up to the CIO to advocate for system security, compliance, and governance to ensure AI safety and risk management. CIOs can get around this quandary by hiring an outside audit firm to monitor AI systems for security, governance, and compliance, regularly reporting to the board and citing any irregularities that must be corrected.

5. Pick the right vendors

Vendors offering AI systems or services must have the same level of commitment to quality AI that your company does. AI vendors should be vetted not only for their products and the data that they use, but for the strength of their support, security, governance, compliance, and ability to recover from a disaster.

CSO has more on what IT leaders should consider here, from an AI security vendor lens: See “Key questions CISOs must ask before adopting AI-enabled cyber solutions.”

A final wrap

Harris/Dataiku surveyed 900 CEOs earlier this year, and 80% said their role was at risk if they didn’t deliver AI results by the end of 2026. Moreover, 57% believed that “insufficient AI explainability could trigger a customer trust or brand credibility crisis.”

Trust in AI has become a major issue for CEOs and their boards. Ensuring AI trust should be at the top of CIO AI to-do lists.