Agentic AI
,
Artificial Intelligence & Machine Learning
,
Next-Generation Technologies & Secure Development
2026 Benchmarking Data Reveals Growing Gap Between AI Ambition and Readiness
In the past year, artificial intelligence move even deeper into business operations, according to Gallagher’s third annual AI Adoption and Risk Survey. Now 63% of organizations say they have fully operationalized or implemented AI within their operations, up from 45% just a year ago – with 82% believing that AI has had a positive impact on the business, and 83% believing it will increase revenues.
See Also: AI Security Risks Rise With Agentic Systems
But AI is also causing some headaches for organizations as they struggle with governance, talent and training, and ROI timelines that can span years.
“In many instances, organizations are still treating AI as a technology initiative rather than a business transformation,” said Ben Warren, managing director of people data, AI and innovation at Gallagher, a global insurance brokerage, risk management and consulting services company.
One way that manifests depends on how a company perceives risk and prepares for risk. The survey found that 93% of organizations say they understand AI risks “quite well” or “very well,” up from 77% in 2024, yet fewer than half have established AI governance frameworks in place. Only 45% have conducted ethical impact assessments, and only 43% have AI incident response plans.
AI errors, hallucinations and misinformation top the list of perceived threats in the survey, cited by 57% of respondents. Second was legal and reputational risk from AI misuse, cited by 56%, followed by data protection and privacy violations at 55%. The survey polled 1,250 organizations of all sizes in the United States, United Kingdom, Canada and Australia.
Complicating the matter, many organizations situate risk governance in IT teams, but it needs to be embedded in operating models and the decision-making process.
“Recognizing risk is very different from operationalizing risk management at scale via a strong governance structure,” Warren said. “CIOs need to ensure clear accountability for AI outcomes, escalation paths when things go wrong, and leadership confidence in managing AI‑related incidents. That requires equipping leaders and embedding AI risk into existing risk management processes.”
The Timeline for ROI: Show Progress
When it comes to communicating AI’s potential to the board, Warren said that CIOs should be tracking adoption depth, early improvements in work quality or decision speed, and capacity building by quarter. This can help show forward momentum to the board as the ROI case builds.
While 63% of companies are actively measuring AI ROI, they estimate it will take an average of 28 months to see that return. Despite the timeline, they’re bullish on AI’s prospects. In fact, 82% say they’re already seeing positive impact, and 83% say they expect AI to boost future revenue.
“Execs and boards often don’t require full financial ROI immediately, but they do need evidence of momentum, rigor and adoption,” he said. “What boards want reassurance on is not just when value will arrive, but whether the organization is building the muscle required to sustain it.”
The Talent Gap Is Now a CIO Problem
More than half of businesses said a shortage of AI-ready talent was the greatest barrier to implementation, and 55% say they are now hiring for AI-specific roles. Another 62% say they’re offering on-the-job AI training, and 30% plan to start.
At the same time, 59% of organizations report workforce reductions or plans to reduce headcount. Some of the decline in headcount is coming from natural attrition or not rehiring for roles that have become more automated.
“This tension exists because AI is changing the shape of work, not simply the volume of it – and many organizations underestimate the change management required to navigate that shift,” Warren said. CIOs will need to work closely with human resources and business leaders to redesign roles across the organization, and invest in training to help employees understand how their roles are evolving alongside AI.
“Businesses that choose to cut capacity without redesigning roles, reskilling people and supporting leaders through the change create long‑term capability gaps and disengagement, presenting tangible risks to their business,” he said.
Most organizations tend to get stuck in the middle of this transformation, scaling tools but not managing culture change. And they fail to invest in the skills and adoption programs needed to use AI effectively at scale.
“As a result, AI remains something people use occasionally, rather than something that can reshape how work gets done and the value that it can provide to the business,” Warren said.
“Most often, the organizations that move through this barrier treat AI as a business transformation, not a digital initiative. They take a phased and pragmatic approach – proving value, learning quickly, and deliberately redesigning the operating model as confidence grows. That’s what separates experimentation from real, sustainable integration.”
Click Here For The Original Source.
