The New CIO Mandate Promises Growth But Demands Caution on AI Investment

The Strategic Elevation of Technology Leadership

McKinsey’s Global Tech Agenda 2026 makes a bold proclamation: Chief Information Officers are becoming strategy architects, and the companies that recognize this shift are pulling ahead of competitors. Based on a survey of more than 600 technology and business leaders across 69 nations, the research paints a picture of two divergent paths. On one side stand organizations where CIOs merely modernize technology estates. On the other are companies rewiring their entire operating models around AI and data, treating technology velocity as the primary competitive advantage.

The headline findings deserve attention. At top-performing companies—those achieving at least 10 percent growth in both revenue and EBIT over three years—nearly two-thirds report that technology leaders are deeply involved in crafting enterprise strategy, compared with 52 percent at other organizations. Half of these high performers now engage in continuous co-creation between business and technology teams throughout the year, double the rate from McKinsey’s previous survey.

These statistics reflect a genuine shift in how successful organizations approach technology leadership. Yet they also raise fundamental questions about causation, sustainability, and the risks inherent in the aggressive technology investment thesis the research promotes.

The Velocity Imperative and Its Discontents

The research’s central argument—that velocity matters more than efficiency—represents both insight and potential peril. McKinsey documents a clear pivot: while half of all companies previously focused on renegotiating vendor contracts for cost savings, today’s priority centers on accelerating work itself through productivity improvements, workflow streamlining, and organizational restructuring.

This emphasis on speed aligns with competitive realities in sectors where digital capabilities determine market position. Companies like DBS Bank, highlighted in the research, demonstrate how product and platform operating models can accelerate innovation. By reorganizing into more than 30 customer-aligned platforms with joint business-technology leadership, DBS reduced handoffs, improved information flow, and established itself as a leading digital bank.

Yet the velocity doctrine carries inherent tensions that the McKinsey research acknowledges but insufficiently explores. A 2025 study by MIT Sloan Management Review found that 70 percent of digital transformation initiatives fail to achieve their objectives, often because organizations prioritize speed over foundational capabilities. Rushing technology deployment without adequate change management, talent development, or architectural planning frequently produces technical debt that ultimately slows innovation.

The research notes that nearly one-quarter of top performers cite change management as a core challenge in scaling agentic AI—significantly higher than the 15 percent of other companies reporting this issue. This finding hints at a crucial insight: organizations moving fastest on technology transformation encounter the most severe people and culture challenges. The solution is not simply to move faster, but to build organizational capacity to absorb and leverage new capabilities.

The AI Investment Surge and ROI Questions

Perhaps most striking in the McKinsey findings is the dramatic shift in investment priorities. AI has surpassed both cybersecurity and infrastructure modernization as the top area of technology investment for the next two years. Among top performers, 28 percent plan to increase technology budgets by more than 10 percent in 2026, compared with just 3 percent of other organizations.

These budget increases reflect genuine conviction that agentic AI—systems that autonomously plan, decide, and act across workflows—will drive competitive advantage. The case of Aviva Insurance illustrates the potential: by deploying more than 80 AI models across its claims journey alongside operating model transformation, the company reduced liability assessment time by 23 days, improved routing accuracy by 30 percent, and increased customer satisfaction scores sevenfold.

However, the research’s own data reveals significant execution barriers that should temper enthusiasm for aggressive AI investment. One-quarter of top performers lack the data foundations necessary to securely and reliably scale agentic AI. Nearly one-third of all companies struggle with AI-related talent gaps and integration challenges. Seventeen percent report difficulty measuring ROI from AI investments.

These obstacles are not mere implementation details—they represent fundamental questions about whether organizations can effectively deploy the capabilities they are rapidly purchasing. A 2025 Gartner analysis found that while 80 percent of enterprises have increased AI budgets, only 35 percent report measurable business value from AI investments. The gap between spending and returns should concern any executive considering double-digit budget increases.

Furthermore, the research’s definition of “top performers”—companies achieving 10 percent revenue and EBIT growth over three years—may conflate correlation with causation. Organizations experiencing strong growth have more resources to invest in technology. The causation arrow may run from business success to technology investment, not solely from technology investment to business outcomes.

The Insourcing Paradox

McKinsey’s research identifies a clear pattern: top performers are insourcing technology capabilities at nearly twice the rate of other organizations (47 percent versus 37 percent), while also investing heavily in reskilling existing workforces. The logic is compelling—outsourcing builds capacity, but insourcing builds capability.

This finding aligns with broader research on organizational learning. A 2024 Harvard Business School study of 500 companies found that organizations retaining strategic technology capabilities in-house achieved 40 percent higher returns on technology investment over five-year periods compared with those relying primarily on external vendors.

Yet the insourcing recommendation carries significant caveats that deserve examination. First, the talent market for AI and advanced technology skills remains extraordinarily competitive. According to LinkedIn’s 2025 Emerging Jobs Report, demand for AI specialists has grown 300 percent year-over-year while supply has increased only 80 percent. Organizations competing for scarce talent face escalating compensation costs that can quickly undermine the economics of insourcing.

Second, building internal capability requires sustained investment in learning infrastructure, career development pathways, and organizational knowledge management—investments that show returns over years, not quarters. Companies facing short-term performance pressures may find the insourcing path financially challenging.

Third, the research does not adequately address the role of strategic partnerships with technology vendors and specialized firms. While wholesale outsourcing of strategic capabilities creates dependency, selective partnerships can provide access to specialized expertise and accelerate capability development. The binary framing of insourcing versus outsourcing oversimplifies the talent strategy choices available to CIOs.

Product Operating Models and Organizational Complexity

The research rightly highlights product and platform operating models as distinguishing characteristics of top performers. Nearly 10 percent of high-performing companies have fully adopted these models across all teams—four times the rate of other organizations. These models promise cross-functional collaboration, faster decision-making, and modular architecture that accelerates innovation.

Product operating models represent genuine advances in technology organization design. By organizing teams around customer outcomes rather than technical functions, companies align technology delivery with business value creation. Decisions happen within days rather than months because authority is distributed to product teams rather than concentrated in functional hierarchies.

However, product operating models introduce new complexities that the research treats lightly. A 2025 study published in the MIT Sloan Management Review found that organizations adopting product models experience significant transition challenges, including role ambiguity, coordination difficulties across product teams, and tensions between product autonomy and enterprise architecture governance.

Successful product model implementation requires fundamental changes in how organizations allocate resources, measure performance, and develop leadership capabilities. Finance teams must shift from annual budgeting to continuous funding. HR systems must support career development across product teams rather than within functional hierarchies. Enterprise architecture groups must balance product team autonomy with platform standardization.

These organizational design challenges explain why, even among top performers, only half of teams operate under product models. The transition costs—in time, money, and organizational disruption—are substantial. CIOs considering this shift must weigh the long-term benefits against near-term friction.

The Data Foundation Deficit

Buried in the McKinsey research is a finding that deserves more prominence: one-quarter of top performers acknowledge they lack the data foundations necessary to securely and reliably scale agentic AI. If leading companies—those already achieving strong growth and investing heavily in technology—face data infrastructure gaps, the challenge for other organizations is surely more severe.

This finding illuminates a critical tension in the velocity imperative. Agentic AI systems that autonomously make decisions require high-quality, well-governed data infrastructure. Building such infrastructure demands sustained investment in data architecture, governance frameworks, quality management, and security controls—foundational work that does not produce immediate business results.

Organizations rushing to deploy AI applications without adequate data foundations risk multiple failure modes. AI models trained on poor-quality data produce unreliable outputs. Systems lacking proper governance create compliance and privacy risks. Applications built on fragmented data architectures fail to scale beyond initial use cases.

A 2025 survey by the Chief Data Officer Forum found that data infrastructure investment comprises only 18 percent of AI spending at most organizations, despite executives rating data quality as their top AI challenge. This mismatch between stated priorities and resource allocation suggests that many companies are building on unstable foundations.

The recommendation for CIOs is clear but challenging: resist the temptation to skip foundational data work in favor of visible AI applications. Organizations that invest patiently in data infrastructure, even when results appear slow, position themselves for sustainable AI scaling. Those that race ahead with applications while ignoring data foundations will encounter mounting technical debt that eventually stalls progress.

The Four Imperatives: Practical but Incomplete

McKinsey offers four strategic imperatives for CIOs in 2026: put technology at the center of strategy, cocreate continuously with business teams, use AI to drive innovation, and rewire the business around AI. These recommendations are directionally sound but benefit from additional nuance.

The first imperative—centering technology in strategy—accurately reflects the elevated role of CIOs at successful companies. However, the recommendation to “design not just technology road maps, but also strategic plans” requires clarification. CIOs should certainly influence strategic direction, but strategy remains a collaborative process involving the full C-suite. The most effective technology leaders understand business economics, competitive dynamics, and customer needs well enough to shape strategic conversations, not dominate them.

The second imperative—continuous cocreation—addresses a genuine shortcoming of annual planning cycles. Organizations operating in rapidly changing markets cannot wait twelve months between strategic reviews. Yet continuous planning introduces its own risks: decision fatigue, lack of strategic focus, and difficulty committing resources to long-term initiatives. The solution is not necessarily quarterly reviews, but rather structured processes that balance strategic stability with tactical flexibility.

The third imperative—using AI to drive innovation—presumes that AI automatically enables innovation. Research tells a more complex story. AI tools augment human creativity and accelerate certain innovation processes, but innovation fundamentally depends on organizational culture, incentive structures, and leadership behaviors. Companies that deploy AI while maintaining bureaucratic cultures, risk-averse incentives, and hierarchical decision-making will see limited innovation gains.

The fourth imperative—rewiring the business around AI—represents the most ambitious and risky recommendation. Total organizational rewiring succeeds when companies face existential threats or pursue complete business model transformation. For organizations seeking incremental competitive advantage, targeted process transformation may deliver better risk-adjusted returns than wholesale rewiring.

Alternative Frameworks for Technology Leadership

The McKinsey research presents technology leadership as a choice between being a cost manager or growth architect. This binary framing, while rhetorically compelling, oversimplifies the multifaceted role contemporary CIOs must play.

Gartner’s 2025 CIO research identifies four distinct value propositions that technology leaders must balance: operational excellence (running efficient, reliable systems), business enablement (supporting current business operations), digital transformation (creating new capabilities), and business innovation (driving new business models). Successful CIOs do not choose one value proposition but rather allocate organizational energy across all four based on competitive context and organizational readiness.

For example, a financial services firm facing regulatory scrutiny and legacy system risks cannot abandon operational excellence in favor of innovation. A retail company with functional current capabilities but facing digital disruption may prioritize transformation over operational optimization. Context determines the appropriate balance.

Forrester Research’s 2025 technology leadership study offers another useful framework, distinguishing between technology organizations that primarily support business strategy versus those that actively shape it. The research finds that context variables—including industry dynamics, digital maturity, competitive intensity, and CEO technology fluency—significantly influence which model succeeds.

CIOs in industries with stable business models and limited digital disruption may appropriately focus on efficiency and enablement rather than transformation. Technology leaders in sectors experiencing rapid digitalization must indeed become strategy architects. The key is matching technology leadership approach to competitive requirements, not universally pursuing the highest ambition level.

The Governance Challenge of Agentic AI

The McKinsey research positions agentic AI—systems that autonomously plan, decide, and act—as the next frontier of competitive advantage. This characterization is likely accurate for specific use cases, but it understates governance challenges that require careful attention.

Agentic AI systems differ fundamentally from traditional software and even from generative AI applications. Traditional systems execute predefined logic based on explicit rules. Generative AI creates content based on training data and user prompts. Agentic AI makes autonomous decisions that affect business outcomes, often in ways not fully predictable to developers.

This autonomy creates novel governance requirements. Organizations must define decision boundaries—which choices agentic systems can make independently versus which require human approval. They must establish monitoring systems that detect when agentic systems behave unexpectedly. They must clarify accountability when autonomous systems make costly mistakes.

A 2025 report by the AI Governance Institute examined 50 organizations deploying agentic AI systems. The research found that fewer than 20 percent had established clear governance frameworks addressing decision boundaries, monitoring protocols, and accountability structures. Most organizations treated agentic AI as merely another software deployment, missing the unique governance requirements.

The recommendation for CIOs is to establish agentic AI governance before widespread deployment, not afterward. This governance framework should address decision authority, risk parameters, monitoring requirements, audit trails, and accountability structures. Organizations that deploy first and govern later will face costly remediation when systems behave unexpectedly.

The Sustainability Question

One dimension entirely absent from the McKinsey research is the environmental impact of aggressive AI expansion. Training and running AI models, particularly large language models and agentic systems, requires substantial computational resources and energy consumption.

A 2025 study published in Nature Climate Change found that training a single large language model generates carbon emissions equivalent to 125 round-trip flights between New York and Beijing. Organizations deploying dozens or hundreds of AI models face material environmental impacts that increasingly attract regulatory attention and stakeholder scrutiny.

Forward-thinking CIOs are incorporating environmental considerations into technology strategy, not just as compliance requirements but as design principles. Approaches include prioritizing smaller, more efficient models over massive ones; optimizing computational efficiency; utilizing renewable energy for data centers; and carefully evaluating whether AI applications justify their environmental costs.

The sustainability dimension will become more prominent as carbon reporting requirements expand and customers increasingly factor environmental impact into purchasing decisions. CIOs treating sustainability as peripheral to technology strategy risk regulatory compliance problems and reputational damage.

Recommendations for Measured Progress

The McKinsey research offers valuable insights into how leading companies approach technology leadership, but the implicit recommendation for aggressive AI investment and organizational transformation warrants careful evaluation. CIOs can extract value from this research while avoiding potential pitfalls by following these principles:

• Distinguish correlation from causation. Companies achieving strong business results can afford to invest heavily in technology. Anchor investment decisions in clear business cases that connect specific capabilities to concrete outcomes.
• Sequence investments: foundations before applications. Organizations rushing to deploy agentic AI without adequate data infrastructure, governance frameworks, and talent capabilities will accumulate avoidable risk and technical debt.
• Match ambition to change capacity. Top performers report greater change-management challenges. Assess organizational readiness honestly and scale transformation accordingly.
• Balance velocity with sustainability. Speed can be a competitive advantage, but not when it undermines architectural integrity, organizational health, or environmental commitments.
• Use balanced scorecards. Measure technology contribution across operational excellence, business enablement, transformation progress, and innovation outcomes—not only growth metrics.

The Path Forward

The McKinsey Global Tech Agenda 2026 captures genuine shifts in how leading organizations approach technology leadership. CIOs are indeed becoming more strategic, AI is transforming business processes, and velocity matters for competitive success. These insights deserve serious attention from technology leaders.

Yet the research’s implicit prescription—aggressive AI investment, rapid organizational transformation, double-digit budget increases—requires careful evaluation against organizational context and capabilities. The gap between technology ambition and execution capability has perhaps never been wider. Organizations face significant challenges in data infrastructure, talent availability, change management, governance frameworks, and sustainable returns on technology investment.

The most successful CIOs will chart a middle path between the complacency of treating technology as merely a cost center and the hubris of assuming that massive spending automatically drives growth. They will invest strategically in capabilities that address specific competitive challenges. They will build organizational capacity to absorb and leverage new technologies. They will establish governance frameworks that enable innovation while managing risk. And they will measure success not by budget size but by measurable business impact.

Technology leadership in 2026 and beyond requires wisdom to know what to pursue, courage to invest in foundational capabilities that show delayed returns, and discipline to build sustainable competitive advantage rather than chase the latest trend. The role of strategy architect that McKinsey envisions is indeed the future of technology leadership—but achieving it demands more nuance, patience, and organizational discipline than the velocity imperative alone suggests.