Asana’s report, published Wednesday, found a gap in enterprise AI agent use: 68% of leaders said they use three or more agents weekly, compared with 36% of workers. The work management platform surveyed more than 4,500 CIOs, AI leaders and workers in the U.S. and U.K. Nearly double the number of agents per week were reported by leaders compared with workers.

Fewer than one-quarter of workers said available agents can handle the tasks where they most need help. About one-third said agents meet some needs but leave significant gaps, while one-fifth said agents mainly handle lower-priority tasks. Nearly half of workers using agents said they spend about 10% of their working time switching tools, choosing agents, managing outputs or resolving conflicts. In addition, 38% said more than one-quarter of agent work is duplicated or overlapping.

Asana Chief Product Officer Arnab Bose said workers face more tools to juggle, outputs to reconcile and uncertainty over responsibility. The report said leaders are more likely than workers to trust agents with information access and autonomous actions. It also said workers’ caution reflects seeing agents act with incomplete context, produce conflicting outputs or escalate decisions needing human judgment. Separately, about one in four enterprise agents run unmonitored. Nearly two-thirds of executives in a recent Workato report expected an agent to cause a public security or compliance incident within the next year and a half. Bose said organizations need people and agents to share a plan, sufficient context and clear responsibility, and urged CIOs to measure completed work rather than agent activity.

Asana’s report, published Wednesday, found a gap in enterprise AI agent use: 68% of leaders said they use three or more agents weekly, compared with 36% of workers. The work management platform surveyed more than 4,500 CIOs, AI leaders and workers in the U.S. and U.K. Nearly double the number of agents per week were reported by leaders compared with workers.

Fewer than one-quarter of workers said available agents can handle the tasks where they most need help. About one-third said agents meet some needs but leave significant gaps, while one-fifth said agents mainly handle lower-priority tasks. Nearly half of workers using agents said they spend about 10% of their working time switching tools, choosing agents, managing outputs or resolving conflicts. In addition, 38% said more than one-quarter of agent work is duplicated or overlapping.

Asana Chief Product Officer Arnab Bose said workers face more tools to juggle, outputs to reconcile and uncertainty over responsibility. The report said leaders are more likely than workers to trust agents with information access and autonomous actions. It also said workers’ caution reflects seeing agents act with incomplete context, produce conflicting outputs or escalate decisions needing human judgment. Separately, about one in four enterprise agents run unmonitored. Nearly two-thirds of executives in a recent Workato report expected an agent to cause a public security or compliance incident within the next year and a half. Bose said organizations need people and agents to share a plan, sufficient context and clear responsibility, and urged CIOs to measure completed work rather than agent activity.