The promises of artificial intelligence continue to grow, and yet, the industry isn’t rushing to adopt the tech as quickly as it wants to. Or as quickly as it thinks it is.

PHD Worldwide, part of Omnicom Media Group (OMG), has released research that looks at adoption, expectations, and apprehensions around AI. The study surveyed more than 700 senior marketers and agency professionals across 10 countries, including Canada. Conducted in partnership with WARC, the research found significant gaps in desire versus adoption, as well as perceived versus actual AI knowledge, in addition to conflicting expectations around the impact of AI.

Right now, the desire to adopt generative AI significantly outpaces current levels of use by the industry. More than one-third of client-side marketers (35%) and agency teams (36%) feel generative AI should be used to a high or extremely high extent, but only 27% of marketers and 26% of agencies are currently using it at this level.

While nearly half of agency and client-side marketers (42%) believe they have advanced generative AI knowledge, their quiz scores reveal a significant discrepancy between perceived and actual knowledge. Only 13.7% of all respondents scored two or more out of five on the generative AI quiz.

Half of agency and client-side marketers are focused on generative AI’s potential for innovation (52% and 51%, respectively), although brand teams tend to prioritize efficiency. Almost half of marketers (48%) prioritize streamlining repetitive tasks and freeing up resources, compared to 39% of agencies.

Lack of technical expertise in AI adoption is a big concern for marketers, with 38% viewing this as a limitation. Meanwhile, 28% of agency teams share this concern.

The research findings also show that the more individuals know about generative AI, the more they agree that it would have a high or extremely high impact on marketing. Specifically, 44% of respondents with above-average knowledge agreed with this statement, compared to 32% of those with below-average knowledge.

Respondents with above-average knowledge were also more likely to agree with the high cost of implementation (43% vs 32% of those with lower levels of knowledge). Only 21% of higher knowledge respondents believe that generative AI will replace action tasks that are currently carried out by agencies within the short term, versus 46% for those with below-average knowledge.