Senior leaders identified as highest AI risk in UK organisations, study claims

Senior leaders are emerging as the highest-risk users of artificial intelligence in UK organisations, according to new research which suggests AI-related business failures are being driven more by leadership behaviour than by technology or junior staff.

The findings come from a survey of more than 2,000 UK tech workers, commissioned by tech and transformation specialist La Fosse for its AI in the Workplace report. The research indicates that C-suite executives combine frequent use of AI with limited safeguards and significant decision-making authority, increasing organisational exposure to risk.

According to the survey, 78 percent of C-suite executives admit to using AI for work they are not trained to do, while 93 percent say they have made AI-informed decisions based on inaccurate data. Four in ten report that these decisions have resulted in serious business impact.

Despite this, confidence at executive level remains high. Seven in ten C-suite leaders describe themselves as very confident in their AI capability. This view is not widely shared across organisations, with only 27 percent of intermediate-level employees saying they trust senior leadership’s AI expertise.

More than half of all tech workers surveyed say AI decisions have been made at their company without the right expertise. Among senior leaders themselves, 65 percent of C-suite executives acknowledge that such decisions occur at the most senior level.

The research also highlights behaviours that increase organisational risk. Nearly three quarters of C-suite executives say they have uploaded confidential company data into AI tools, compared with 42 percent of entry-level staff and 35 percent of intermediate employees. Senior leaders are also more likely than other groups to rely on AI outputs for high-stakes decisions while operating with limited oversight.

The consequences of these practices appear to be concentrated at the top of organisations. While 40 percent of C-suite executives report serious business impact from AI errors, the figure falls to 32 percent among entry-level staff and 11 percent among intermediate employees. La Fosse suggests that greater autonomy, time pressure and responsibility at senior levels increase exposure to risk faster than AI capability is developing.

Ollie Whiting, chief executive of La Fosse, said the people with the greatest autonomy over AI were also the most exposed to its risks. He said confidence and speed were often masking gaps in governance, skills and accountability, and warned that organisations needed to ensure leaders had the right expertise before AI-related errors escalated into business-critical failures.

The study also identifies a widening trust gap between leadership and the wider workforce. While 70 percent of C-suite executives rate themselves as very confident in their AI capability, fewer than half of directors and senior managers share that view, falling to 33 percent among entry-level staff. Despite this confidence, 80 percent of C-suite respondents say a dedicated AI specialist is needed at board level.

The research also found that half of tech workers expect AI to lead to job losses at their organisation within the next three years, suggesting a workforce preparing for significant disruption alongside growing concerns about how AI decisions are being governed.