AI agents could add USD $450 billion, but UK trust declines

Research from Capgemini indicates that AI agents could add up to USD $450 billion in economic value globally by 2028, though the rate of full deployment and trust in these systems among UK organisations remains low.

The study, conducted by the Capgemini Research Institute, surveyed 1,500 business leaders from 14 countries, including 135 executives from the UK. It found that while 47% of UK businesses are running pilots or experimenting with AI agents, only 1% have fully implemented the technology at scale. Nineteen percent have achieved partial implementation to date. The figures represent a notable increase considering that global adoption rates stood at 4% just a year earlier.

Organisations' expectations for AI agents are evolving. By 2028, more than a third (38%) of UK business leaders anticipate AI agents acting as autonomous members in human-supervised teams, with 21% expecting agents to coordinate tasks within those teams and 12% predicting that agents will supervise other AI agents.

Despite anticipated growth, trust in autonomous AI systems has declined. In the UK, confidence in autonomous AI agents for enterprise use fell by 18% over the past year, dropping from 50% in 2024 to 32% this year. Across the board, 64% of UK respondents say they do not trust autonomous AI agents.

The findings highlight a gap between executive ambition and organisational readiness. Almost all (93%) business leaders see scaling AI agents within the next year as key to competitive advantage, yet close to half lack a clear strategy for adopting these systems.

Benefits and hesitations

Capgemini's global findings reinforce the importance of human oversight and collaboration alongside the adoption of AI. The report notes that effective teaming between humans and AI agents could boost human engagement in high-value tasks by 65% globally, with increases in creativity (53%) and employee satisfaction (49%).

Despite these projected benefits, 61% of UK business leaders report that employees are concerned about the impact of AI on their jobs and career prospects. At present, only a third (34%) of UK businesses say they understand where best to employ AI agents, while more than half (52%) report insufficient advanced data infrastructure for managing AI solutions.

Franck Greverie, Chief Portfolio & Technology Officer, Head of Global Business Lines, and Group Executive Board Member at Capgemini, stressed that achieving economic gains from AI agents will require broad organisational change. He stated:

"The economic potential of AI agents is significant but realising this value depends on more than just the technology, it requires a comprehensive and strategic transformation across people, processes and systems. To succeed, organisations must remain focused on outcomes, reimagining their processes with an AI-first mindset. Central to this transformation is the need to build trust in AI by ensuring it is developed responsibly, with ethics and safety baked in from the outset. It also means reshaping organisations to support effective human-AI chemistry, creating the right conditions for these systems to enhance human judgment and help deliver superior business outcomes."

Transparency and trust

The report shows that organisations are prioritising transparency to address the decline in trust. While only 40% of businesses currently trust AI agents to manage tasks autonomously, among those in implementation phases, 47% report above-average trust levels compared to 37% of those still in exploration.

Concerns over data privacy, algorithmic bias, and explainability remain prevalent. Privacy was cited as the primary concern for 51% of organisations, with only 34% actively working to mitigate these risks. According to the findings, fewer than one in five organisations report a high level of data-readiness, and 80% lack mature AI infrastructure needed for large-scale deployment.

Looking ahead

Over the next 12 months, more than 60% of organisations expect to establish human-agent teams in which AI agents take on subordinate or capability-enhancing roles. Around 70% of businesses anticipate that AI agents will drive organisational restructuring, prompting considerations around team composition and workflow redesign.

At present, about 15% of business processes are reported to be semi- or fully autonomous, a figure the report expects to rise to 25% by 2028. Initial adoption is concentrated in customer service, IT, and sales, with expansion into operations, research and development, and marketing forecasted over the coming years.

Companies that have already scaled their use of AI agents stand to realise higher economic gains, with projections of average returns of USD $382 million over three years, compared to USD $76 million for those still at the early stages of implementation.

The report recommends that organisations focus on redesigning processes, reshaping business models, and finding the right balance between agent autonomy and human involvement to ensure successful adoption and value realisation from agentic AI.