The study highlights a significant rise in the proportion of total AI budgets allocated to generative AI compared to the previous year.

Despite this optimistic outlook, organizations face notable challenges in developing and implementing Gen AI tools. Regulatory compliance has emerged as the top barrier, cited by 38% of organizations, a 10-percentage point increase from last year. Difficulties in managing risks (32%) and implementation (27%) follow closely. Addressing these governance foundations will require perseverance, as nearly seven out of ten organizations (69%) expect it will take over a year to fully implement a governance strategy.

C-suite leaders (CxOs) exhibit greater optimism regarding Gen AI implementations within their organizations compared to other study participants, with 21% of CxOs reporting that Gen AI is already transforming their organization, versus only 8% of non-C-suite respondents. CxOs are also less concerned about challenges such as trust, risk management, governance, and regulatory compliance.

Alexandru Reff, Country Managing Partner, Deloitte Romania and Moldova, said: "The report shows that most companies are transforming at the speed of organizational change, not at the speed of technology. It recommends that leaders should stay curious and challenge the orthodoxies of their organizations. While uncertainty about specific outcomes and their timing is unavoidable, executives should focus on what they can control - namely, organizational readiness, particularly in areas such as data, risk management, governance, regulatory compliance and talent".

The study also indicates that agentic AI (52%) and multiagent systems (45%) are perceived as the most promising technologies for future Gen AI development. While many organizations are exploring autonomous agent development, over two-thirds of respondents believe that 30% or fewer of their current Gen AI experiments will be fully scaled within the next three to six months. Nevertheless, 74% of organizations with advanced Gen AI initiatives report that their projects are meeting or exceeding ROI expectations. 

IT (28%) remains the leading function for advanced Gen AI initiatives, with other areas such as marketing, operations, and cybersecurity also seeing significant application.