What a year of AI, quantum, cyber risk, and infrastructure constraints reveals about the next era of enterprise leadership
In the past year, enterprise technology risk has become more systemic, more physical and more entangled with policy and economics. It’s no longer possible to plan technology strategy in isolation – it must now account for the coverage of capital strategy, geopolitical strategy, workforce strategy, sustainability strategy and operational resilience.
The next twelve months will be less about discovering entirely new technologies and more about absorbing the consequences of the technologies that have already been deployed, writes Kris Lovejoy, Chief Strategy Officer, Kyndryl. In this new era, enterprises will discover that the hardest part of technology transformation is not invention; it is control.
Key takeaways
AI is moving from experimentation to economic discipline.
The pilot phase of generative AI is ending, and it is now becoming a run-rate expense, an operational dependency, and in some cases, a source of budget volatility. Agentic AI changes the nature of consumption, as a single user can trigger dozens or hundreds of model operations in a single business interaction. In addition to token consumption, conversations about open-weight models raise questions about whether every AI engagement must run through a “metered” service.
The enterprise is creating an ungoverned AI estate.
Unlike shadow IT, AI agents are full actors within the enterprise environment who can retrieve data, initiate transactions, generate code and more. Yet in recent research from Kyndryl, just 27% of executives told us they had a central registry and monitoring system for all AI operating in their environments. Without visibility into what agents exist, who owns them, what they can access and more, enterprises will struggle to answer essential and basic governance questions.
Digital growth is becoming constrained by physical resources.
Physical constraints, including power, water and community acceptance, are increasingly limiting AI adoption. This shift will change enterprise technology procurement as buyers begin asking cloud and AI providers the questions once reserved for utility providers, environmental teams and local governments.
Quantum is becoming a governance problem before becoming a compute advantage.
Quantum’s near-term impact begins with governance. Organizations must prepare for a future in which cryptographic assumptions change, optimization methods improve, and industry-specific quantum use cases begin moving from research partnerships toward early workflow integration. Waiting for a dramatic breakthrough could leave enterprises unprepared and cost them the opportunity to gain first-mover advantage.
