EY published AI Sentiment Report 2026, which shows that, while public debate regarding artificial intelligence (AI) continues to focus on trust and risks, real-world behavior indicates a different reality. Consumers are increasingly integrating AI into their daily decision-making and are already allowing AI systems to act on their behalf.
In the last six months alone, more than eight out of ten respondents (84%) have used artificial intelligence, and globally, 16%have stated that they use AI systems that operate without human intervention. This suggests that the use of artificial intelligence is moving beyond the initial stage of experimentation, and autonomy is no longer a future scenario but an evolving reality.
The second annual edition of the study EY AI Sentiment Report surveyed over 18,000 individuals aged 18 and older, across 23 markets, with the goal of understanding how people have perceived and used artificial intelligence in the last six months. This year's findings are crystal clear: AI adoption is outpacing distrust.
Raj Sharma, EY Global Managing Partner – Growth & Innovation, a declarat: „Concluziile cercetării indică faptul că o minoritate în creștere deleagă deja decizii către inteligența artificială, în timp ce mult mai mulți o utilizează ca pe un instrument de sprijin în viața de zi cu zi. Pentru liderii de companii, această situație creează oportunități, dar și responsabilități. Pe măsură ce inteligența artificială autonomă începe să se extindă, organizațiile trebuie să modeleze acest proces într-un mod deliberat, apăsând pedala de accelerație acolo unde există deja încredere și valoare și cu precauție acolo unde sunt încă necesare claritate, măsuri de protecție sau confort”.
From assistance to delegation
The report shows that the rapid adoption of artificial intelligence was driven less by an increase in user confidence in the technology and more by familiarity resulting from everyday use in low-stakes activities such as route planning, customer assistance, travel planning, and content recommendations, which have become integrated into daily routines.
Among the respondents:
- 9% used autonomous vehicles or taxis
- 10% have used AI agents to purchase products on their behalf
- % allow artificial intelligence to automatically complete shopping carts or manage banking operations.
Even among those who have not yet used autonomous artificial intelligence, the degree of openness is high. More than a third of respondents stated that they would prefer artificial intelligence to automatically apply discounts at checkout (36%) or to resolve customer service issues without their involvement (34%), and a significant proportion are open to artificial intelligence taking the initiative in higher-stakes scenarios such as home security (30%) or scheduling appointments (21%).
The pace of adoption is increasing faster than trust in artificial intelligence.
Although usage is expanding, confidence in how artificial intelligence is controlled has not kept pace. Across all markets, people continue to express fears about security, control, accountability, and authenticity; however, these concerns are not slowing the pace of adoption. Instead, they are shaping expectations about how autonomy should be introduced.
The study found that:
- 66% of respondents are concerned about the possibility of AI systems being illegally accessed or affected by security breaches
- almost seven out of ten respondents (66%) believe human oversight remains essential
- 73% of respondents, nearly three-quarters, fear they will no longer be able to distinguish between real and AI-generated content.
Joe Depa, EY Global Chief Innovation Officer, added: „Trust will be the factor that determines the long-term winners in the AI economy, but currently adoption is advancing faster than the level of trust. Organizations must earn this trust through positive daily experiences, supported by reliable data, clear safeguards, and accountability, to bridge the gap between behavior and perception.”.
Pioneer markets offer a first signal
Although early adopters appear in all markets, the pace of adoption is not uniform. Research has identified eight „pioneer” markets that are further along in their AI journey and where its use is more widespread, frequent, and deeply integrated into everyday life. These markets are India, Mainland China, Brazil, Mexico, Saudi Arabia, the UAE, Hong Kong, and South Korea. Cumulatively, 94%of people in these markets reported using AI, with nearly one in four (24% ) already using autonomous AI.
The other markets were divided into „transitioning markets,” where general adoption has been slower but usage and interest are beginning to grow, and „lagging markets,” which exhibit greater reluctance, lower perceived relevance, and more selective usage. These are 12% and 15% behind the transitioning markets in overall AI adoption, and 11% and 13% behind them in autonomous AI usage, respectively.
„In Romania, we see AI adoption primarily focused on increasing personal efficiency, but still limited when it comes to the profound transformation of business processes. AI governance is in its infancy, and many organizations remain in the realm of experiments and pilot projects that fail to scale. Although numerous companies have taken the first steps, the lack of a clear direction regarding the next stages generates some frustration about the real benefits AI can deliver. In this context, we are witnessing a clear increase in interest in educating boards of directors and executive directors, who need to understand the impact of AI on organizational strategy, as well as the concrete steps required for adopting agentic AI systems,” stated Aurelia Costache, Partner responsible for AI & Data at EY Romania.
