Artificial intelligence has been a dominant topic in public debate for years, particularly in the context of business value. Despite reports of ever-new global investments and the growing availability of various tools, many entrepreneurs and managers continue to ask themselves how profitable and feasible AI implementation really is in Polish business conditions. Salesforce decided to address these doubts by conducting the AI Readiness Index study, covering 200 large, medium-sized and small Polish companies. The results were presented during this year’s edition of Agentforce World Tour Warsaw.
Widespread AI adoption: who, how much and why?
Poland ranks highly in global comparisons of openness to the implementation of technological innovation, serving as a testing ground for many solutions in banking, sales and automation. The situation is no different in the case of artificial intelligence. As many as 44% of Polish companies declare that AI is a high or even key priority in their implementation plans. Only 25% of the surveyed companies have no plans related to artificial intelligence.
At the same time, nearly half of the companies that have decided to adopt intelligent tools — 46% — have been using them at full scale for more than a year. These results point to a clear trend: artificial intelligence is no longer merely an experiment, but is becoming a permanent part of the strategies of Polish companies. The question is therefore no longer “whether organisations in Poland are using AI”, but rather “what they are using it for and with what results”.
LLM-based tools are by far the most popular among the surveyed companies, used by around 90% of organisations that already apply AI. More than one in three companies — 37% — use custom solutions designed specifically for their needs. One in four — 25% — use automation tools or copilots, while 11% use generative tools.
The implementation of artificial intelligence has a measurable impact on improving internal organisational work. Companies declare that, thanks to AI, they have managed to automate an average of 21% of their processes. At the same time, three in four companies that are testing, implementing or using AI solutions — 75% — report that their employees have generally responded positively to the proposed changes. Among companies that have defined KPIs for artificial intelligence — 19% — as many as 60% report a reduction in the time needed to complete tasks, while nearly half — 49% — observe that employees spend less time on repetitive and tedious activities.
What, then, is the main motivation for companies turning to artificial intelligence? The most frequently cited reason is the improvement and streamlining of processes, mentioned by 73% of respondents. Nearly one in three companies — 31% — wants to use AI to reduce costs, while one in four — 24% — sees artificial intelligence as an integral part of a broader strategy.
For balance, respondents were also asked about arguments against AI implementation. Most companies are concerned about high costs and insufficient budgets — 30% — while one in four — 27% — is worried about data security. Nearly one in five organisations — 19% — faces management-related challenges, including reluctance among senior executives to embrace change, a lack of clear AI use cases or insufficient time to plan and carry out implementation.
Half of the Polish market is ready for AI — what about the rest?
In total, as many as 75% of the surveyed companies have, to varying degrees, implemented, are implementing, are testing or are seeking information about AI solutions. However, declarations and willingness must be reflected in real “readiness” — namely appropriate data architecture, technological infrastructure and team competencies. How is the Polish market performing in this respect?
According to the study, the market is almost perfectly divided in half.
The AI-ready group consists of companies that approach implementation professionally: 53% offer employees training in the use of artificial intelligence, while 47% have a dedicated team or coordinator overseeing AI projects. The technological layer and IT infrastructure are no less important, as they allow AI to be effectively integrated with the systems already used by the company. More than half of the surveyed companies — 51% — have such capabilities.
“Artificial intelligence is a growth engine that offers the chance to achieve unmatched performance. Every machine, however, needs a skilled driver, because without that person’s abilities, even with the greatest power and potential, it remains only a tool. Polish companies that are consciously preparing for AI transformation understand this very well, investing in the people who will take the wheel,” comments Krzysztof Augustynowicz, Regional Vice President Central & Eastern Europe at Salesforce.
The most common form of support for employees is access to educational materials and training — 46% — but establishing clear rules, limitations and procedures for the use of AI tools is also important, cited by 41% of companies. The effectiveness of employee training is also confirmed by the experience of thousands of users of the Salesforce Trailhead platform, while the study’s findings further confirm that the Polish market is not lagging behind global competitors.
However, when looking at the other side of the spectrum, the situation appears much darker. Nearly half of Polish companies do not have adequate technological, managerial or training resources. At the same time, more than four in ten organisations — 43% — do not plan to appoint a senior executive to manage AI transformation in the near future. A significant part of the market therefore remains in a maturing phase, and the lack of responsible leadership in AI implementation may produce the opposite of the intended effect: instead of productivity, it may generate organisational chaos.
Agentic AI: a sector with potential, but still in its infancy
One of the most advanced areas of contemporary artificial intelligence is agentic AI, which offers autonomous, capable agents able to support almost any business process. In Poland, however, this sector still has a long way to go. Tools based on agentic artificial intelligence are currently used by just over one in ten companies — 12%.
Among companies that do use such tools, the most popular functions are those related to integration between applications — 61% — and the automatic modification of data in CRM, ERP and similar systems — also 61%.
Although today’s AI agents, such as Agentforce, can perform a variety of tasks without human intervention, full autonomy is still not the norm. More than two thirds of companies — 70% — say they still need to monitor an agent’s progress and detect errors.
The next five years: heading towards the Agentic Enterprise
According to the predictions of surveyed business leaders, more than one in five companies — 23% — expect AI to be absolutely critical to their survival in the market within five years. Nearly half — 49% — believe artificial intelligence will be very important. Only 12% of companies still maintain that AI will have only marginal significance.
The results of the AI Readiness Index show that the Polish market is currently at a transitional stage: between experimentation and real transformation. Competitive advantage will no longer be determined merely by access to technology, but by an organisation’s ability to organise its data, prepare its people and develop a responsible model for implementing AI. Artificial intelligence will not replace companies. But companies that fail to learn how to use it may be replaced by those that gain an advantage through transformation.
“Polish companies need to rethink their infrastructure in order to fully harness the potential of agentic AI and minimise risk,” explains Krzysztof Augustynowicz. “Large language models do not possess specific business knowledge, do not operate on the basis of cause-and-effect relationships and raise concerns about transparency. They can also generate hallucinations, which is completely unacceptable for organisations. To move beyond the ‘pilot’ phase and deploy agents at enterprise scale, models must be supplemented with a system that provides them with context — corporate data — a workspace system consisting of applications supporting business processes, a dedicated AI agent system and an engagement system connecting agents, customers and employees,” he adds.
Salesforce combines these four systems within its Agentic Enterprise Architecture. This architecture is based on Data 360 as a context system for the effective use of enterprise data; Customer 360 as a workspace system built on 27 years of innovation in enterprise applications; Agentforce as a system for creating, testing, deploying, monitoring and scaling agents; and digital and voice channels, as well as Slack, as the application where a new way of working is developing — one that connects people, agents and applications. The architecture is open at every level and can be integrated with all of a customer’s business systems.
About the AI Readiness Index
The AI Readiness Index was developed by SW Research and ITBC Communication.
The study used computer-assisted telephone interviewing (CATI). It covered 200 surveys conducted among companies in Poland: small companies (N=50), medium-sized companies (N=100) and large companies (N=50). Respondents were decision-makers: business owners, management board members, CTOs, CIOs and senior executives.
The study was conducted between 17 and 27 April 2026.
Source: Managerplus.pl
