Dive Brief:
- Nearly 9 in 10 enterprise leaders are identifying roadblocks to agentic AI adoption, according to the latest EY U.S. AI Pulse Survey published last week. The professional services firm commissioned a third party to survey 500 U.S.-based business decision makers across a variety of industries.
- Key adoption barriers include cybersecurity concerns, data privacy issues, lack of regulation and company policy gaps. Those surveyed said they believe around 3 in 5 senior leaders in their industry — and more than half in their own organizations — don’t understand the technology’s benefits.
- The majority of decision makers also worry about employee buy-in. More than 60% of leaders agree that fear of replacement will stifle agentic AI adoption. Employee pushback is leading to an investment in training, with 64% anticipating more upskilling over the next year compared with just 49% who said the same last year.
Dive Insight:
Enterprises are under pressure to perform as market volatility persists, and many are placing confidence in technology projects, particularly AI, to boost productivity and drive growth. While enterprise attention may be on AI agents this year, earlier forms of automation still reign supreme for executives.
“That classic AI, some generative AI, is where most organizations are focusing their efforts today, and the good news is they’re getting a good return on their investment,” Traci Gusher, AI and data leader at EY Americas, told CIO Dive.
In EY’s survey, 97% of senior leaders investing in AI have experienced positive ROI, with those allocating 5% or more of their budget seeing greater returns. Business decision makers often point to employee productivity and operational efficiency as key metrics.
AI agent adoption and its ROI are much more novel, but analysts expect enterprises that dabble in agentic AI now are gaining a head start.
“There’s a few different reasons why we’re seeing hesitation” on agentic AI adoption, Gusher said, pointing to business disruption and other risks or roadblocks. “Those organizations that hold back are going to regret that.”
Tweaking processes to align with regulation changes or iterating on projects as new capabilities emerge is an easier lift than starting from scratch, Gusher said.
“The paralysis of the unknown is the bigger danger than moving too fast and then having to go back and modify,” Gusher said.
Most enterprises, however, have work to do before jumping on the agentic AI bandwagon.
“While the allure of autonomous systems is strong, our findings underscore the importance of a deliberate and human-centric approach to agentic AI,” Whitt Butler, vice chair, consulting at EY Americas, said in a statement. “The future of work will be shaped by how well organizations prepare their people, embed responsible governance and align AI capabilities with real business outcomes.”
Early adopters are seeing value from rolling out the technology. Out of those organizations that have deployed the technology, most are using it to assist and manage processes, tasks and workflows, according to the EY survey.
