CCI staff share recent surveys, reports and analysis on risk, compliance, governance, infosec and leadership issues. Share details of your survey with us: editor@corporatecomplianceinsights.com.
Nearly half of organizations have bypassed AI governance
Just under half (47%) of organizations have circumvented AI policies for urgent deployments, a survey by EY concluded.
The survey of 202 US-based senior AI decision-makers at companies with at least $1 billion in revenue found that almost half say their firms have bypassed AI governance processes when they needed to quickly deploy the technology. Nearly all (98%) of these companies have formal AI policies in place.
But those policies are coming up short of technological advancements, according to the survey. About half (49%) of respondents whose organizations use agentic AI said their governance framework hasn’t been updated to cover agentic tools, and nearly two in five (39%) of agentic AI-using companies said accountability for maintaining or monitoring machine agents after deployment is undefined.
However, companies are catching up, the survey concluded. Almost all (98%) reported that their organizations have annually conducted AI assurance reviews. After these reviews, a quarter of companies fully stopped some of the AI they were running. Another 64% significantly modified AI after a review.
The survey found that the most common AI problems reported in assurance reviews were related to data quality (57%), AI model drift (48%) and shadow AI (39%).
Risks are harder to predict for 2 in 3 audit leaders
Nearly two-thirds of audit leaders are finding it more difficult to see risks before they hurt, according to a survey by Gartner, which found that 64% say it’s harder now to spot risks before they have a material impact.
The survey of 108 audit leaders suggested that fast AI adoption, rapidly evolving regulation and geopolitical uncertainty are overwhelming governance, controls and risk management practices.
While audit leaders are having trouble predicting risks, business leaders are struggling to translate assurance insights into actions. Only 30% of leaders said their ability to manage risks was heavily influenced by insights from audit, compliance and enterprise risk management (ERM).
“This environment has raised the bar on what is required of business leaders,” Tegan Gebert, vice president in the firm’s assurance practice, said at a Gartner event. “What it takes for them to fulfil their risk responsibilities is not only harder to achieve, but also even more critical to get right.”
US businesses see less ROI on AI than others
A much lower percentage of US businesses are being blown away by AI’s return on investment than European and Asia-Pacific companies, a survey by IDC and Expereo found.
The survey of 800 multinational enterprises with 500 or more employees concluded that only 15% of US businesses reported that ROI on AI exceeded expectations compared to 40% of companies in the Asia-Pacific region. Globally, 38% of organizations said that AI only partially met ROI expectations, while 37% said the technology broadly met return expectations.
European companies are the most likely to walk AI back, with 22% reporting that they’re down-scaling after unmet expectations; that rate was 10% in the US and just 6% in Asia-Pacific.
The survey also showed that adoption of AI may not be as widespread as it can appear. Almost two in three organizations (62%) described their AI use as “limited,” while 30% said their AI use was “extensive” or “transformative.”
Of those investing in AI, fear of falling behind is a motivating factor, but that rate differs by region, the survey found. In the US only 10% of leaders are investing in AI because of fear of lagging behind, but, in Asia-Pacific, 37% are taking up AI for fear of losing the pack.
Almost 50% of workers find companies’ AI training lacking
Nearly half of employees don’t believe their organizations provide decent training for the safe utilization of AI, according to a survey by TrustedTech.
The survey of 2,001 US and UK employees across industries found that 44% of workers reported their organizations lack adequate training on using AI safely and securely, and that percentage rises among company leaders, of whom 53% gave the same assessment of AI training.
Mostly, employees and executives aren’t receiving most of their AI education through organization training, the survey found. Nearly a third of decision-makers (30%) reported they were self-taught on AI, while even more workers (41%) reported learning on their own. Nearly a quarter of decision-makers (24%) said they got their AI skills from online resources like YouTube and blogs with nearly 14% of workers saying they learned from those same self-accessed, online sources. For 32% of leaders and about 15% of workers, they reported their AI skills came from employer training.
The survey also concluded that a “contradiction” exists among IT and telecoms workers. They feel the most secure using AI, with 87% reporting they’re confident in using the technology effectively, yet 68% are concerned about shadow AI, or unauthorized AI use.
“IT and telecoms should theoretically be the industry best prepared for AI, but our data shows that confidence and readiness are two different things,” Julian Hamood, founder of TrustedTech, said in a news release about the survey. “You can have a workforce that knows how to use AI and still lack the guardrails for safe adoption. That gap is exactly where shadow AI lives in every industry.”
