AI investment hits new heights — and CIOs are on the hook

The clock is ticking.

Artificial intelligence has transitioned from its introduction as a shiny new toy to becoming a cornerstone of companies’ operations.

A recent survey by McKinsey & Company found that AI has overtaken cybersecurity and infrastructure modernization as companies’ top area of technology investment for the next two years, underscoring how central the technology has become to corporate growth plans

CIOs, the firm said, are increasingly being pulled into enterprise strategy and are now expected to drive measurable business results.

“Technology expertise has become strategy expertise,” McKinsey said.

More than half of the respondent companies identified AI as a priority investment, reflecting what the firm described as a broader shift in how executives view the technology.

“AI has become a business imperative,” the firm said.

The spending reflects that imperative. 

CIOs concerned about budgets

Industry forecasts suggest AI capital expenditure is projected to hit unparalleled levels this year, with tech giants Amazon (AMZN), Microsoft (MSFT), Alphabet (GOOG), and Meta Platforms (META) forecast to invest roughly $650 billion to $700 billion collectively.

And with the expense comes expectations, as companies want to see some return on these massive investments. CIOs say they are feeling the heat.

More on AI stocks:

• Morgan Stanley sets jaw-dropping Micron price target after event
• Bank of America updates Palantir stock forecast after private meeting
• Morgan Stanley drops eye-popping Broadcom price target

A separate survey by enterprise AI platform provider Dataiku found 71% of CIOs say it’s likely their AI budget will be cut or frozen if targets aren’t met by the end of the first half of 2026.

Failed initiatives, the company said, frequently lead to significant financial losses and can put the jobs of CIOs and other senior technology leaders at risk.

“AI is everywhere in large enterprises. Results are not,” Dataiku said. “AI has entered a new accountability era, in which budgets, compensation and executive credibility are increasingly tied to provable outcomes rather than ambition.”

Beyond budget scrutiny, governance and oversight are emerging as major pain points.

Seventy percent of CIOs surveyed expect new audit or explainability requirements for AI systems within the next 12 months, while 85% say gaps in traceability or explainability have already delayed or prevented AI projects from reaching production.

“CIOs are moving from experimentation into accountability faster than most organizations expected,” Florian Douetteau, co-founder and chief executive of Dataiku, said.

Despite years of experimentation, many CIOs now say their organizations’ AI foundations are weaker than anticipated.

Nearly three-quarters of respondents said they regret at least one major AI vendor or platform decision made during the past 18 months, and 62% said their chief executive has directly questioned those choices. Almost one-third said they have repeatedly been asked to justify AI outcomes they could not fully explain.

Dangers of shadow AI

CIOs are also bracing for the downside if the AI market contracts or an “AI bubble” bursts. Most respondents expect major disruption to their company, and 60% say their own job would be at high risk. 

Governance challenges are being compounded by the rapid spread of so-called “shadow AI” — the use of unauthorized tools by employees inside organizations.

Related: BofA revamps Applied Materials stock price target after CEO message

More than half of CIOs surveyed said they have already discovered unsanctioned AI use in their companies. Eighty-two percent said employees are creating AI agents and applications faster than IT teams can govern them, and 89% believe uncontrolled AI access will create significant technical debt.

As AI agents increasingly influence business-critical workflows, the visibility gap is widening. While 87% of CIOs said AI agents are already embedded in critical operations, only 25% reported having real-time visibility into all AI agents running in production.

Shadow AI can expose sensitive corporate data, create regulatory and privacy risks, and introduce bias into hiring, promotion, and compensation decisions, according to the Society for Human Resource Management.

The broader problem is not limited to AI alone. A 2023 report by Gartner projected that 75% of employees will use some form of shadow IT by 2027, up from 41% in 2022.

This is being driven largely by the democratization of technology, the rise of remote work, and employees’ reliance on unauthorized software-as-a-service (SaaS) applications and generative AI (GenAI) tools to boost productivity.

Employees themselves frequently acknowledge the tension between corporate restrictions and the practical benefits of generative AI tools.

“We’ve banned it, but everyone keeps doing it anyway,” one user wrote on Reddit, referring to the use of personal AI accounts for workplace tasks. “I realize that we have to give our tool to our people so they stop using theirs.”

In its report, Dataiku outlined seven decisions it said would determine whether AI becomes a long-term competitive advantage or a growing liability for companies.

“AI must scale beyond IT without surrendering control,” the company said. “CIOs overwhelmingly agree that broader access to AI is essential — but only if governance, monitoring and guardrails are built in from the start.”

Related: Goldman Sachs quietly hands the boring stuff to Claude Al