[Download the report or read the summary below]
AI is being used. It is far less often changing business performance
In most organizations surveyed, AI remains at the stage of learning, prototyping, or isolated improvements. Tools are being applied to selected tasks and parts of processes, but they are far less frequently transforming how the entire business operates.
The report explores what happens between gaining access to technology and generating business value: how companies implement AI, what they consider success, where barriers arise, and what they need to scale up.
Three figures that reveal the scale of the challenge
-
7% full-scale adoption remains the exception
Only a small number of organizations report integrating AI into core processes and achieving measurable financial results. Most are still at an earlier stage on the journey from initial use cases to lasting value.
-
67,5% of companies add AI to existing ways of working
The most common approach is to add AI to an existing process or automate individual tasks. The technology speeds up operations, but does not necessarily lead to a redesign of how work is done.
-
82,8% would allocate additional budget to the foundations
Companies most often point to skills, data, and strategy. This suggests that the next step may not be buying another tool, but preparing the organization to select and implement the right initiatives.
What lies behind these figures?
The full picture is more complex. Organizations differ in how they measure outcomes, assess barriers, evaluate implementation maturity, and define investment objectives. Executive boards, risk teams, and those managing day-to-day operations each view the challenge differently.
The report also highlights differences between medium-sized and large companies, the extent of unauthorized AI use, industry-specific investment objectives, and the decisions that must be made before moving from pilots to large-scale deployment.
Chief Economist, Grant Thornton Poland
Most companies are still exploring the real capabilities of this new type of technology. The demand for strong foundations is therefore entirely rational: targeted implementation of what works, education, and prototyping. At the same time, the organizations with the greatest chance of significant success will be those that are first to fully understand the new technology, implement it effectively at scale, and track key metrics and KPIs throughout the process.
Choose a program tailored to your organization’s situation
The report helps identify where AI adoption stalls. The next step depends on the challenge your organization needs to address today.
We do not know where AI can create value or whether we are ready for our first investment
We identify opportunities for value creation and assess the readiness of data, technology, processes, people, risk management, and accountability. The outcome is a well-founded decision on the first AI initiative: AI Value and Implementation Readiness Assessment Program
We have ideas or pilot projects, but they do not progress to implementation and everyday use
We organize the initiative portfolio, select the use case with the greatest potential, develop a business case, and guide the solution through implementation and adoption: AI Implementation and Adoption Program
We have initial implementations, but every new initiative starts from scratch
We establish a shared operating model, standards, a portfolio management approach, and a value measurement system so that future initiatives can be launched faster and more predictably: AI Scaling Program
The tools are available, but ways of working and responsibilities remain unchanged
We define how tasks should be divided between people and AI, redesign roles and responsibilities, and identify the skills needed to use these solutions in everyday work: AI Work Organization Program
AI solutions are being developed faster than the rules for approving and overseeing them
We create an initiative register, risk classification, decision-making authority, and control gates to manage risk without blocking valuable use cases: AI Governance Program
Senior Business Designer, EDISONDA | Grant Thornton Poland
The organizations that stand to gain the most are those that, instead of asking where AI can be added to an existing process, consider what that process, service, or product would look like if designed from scratch today. Competitive advantage will not belong to companies that implement the most AI tools, but to those that stop optimizing the past and start redesigning how they operate.
