Organisations report a four-fold increase in the deployment of generative AI since 2023

• The vast majority (80%) have increased their year-on-year investment in generative AI
• Three quarters agree it is helping drive revenue and innovation
• Almost all organisations (97%) allow employees to use generative AI in some capacity

Early adopters have already begun reaping benefits, ranging from improved operational efficiency to enhanced customer experience and increased sales. For example, on average, organisations realised a 6.7% improvement in customer engagement and satisfaction over the past year in the areas in which the technology has been piloted or deployed.

“Generative AI is starting to transform business and organisations are already seeing concrete growth in revenue, whilst also accelerating innovation. As a result, rather than solely focusing on cost optimisation, businesses are actively exploring new avenues to leverage its capabilities and drive value creation,” comments Pascal Brier, Chief Innovation Officer at Capgemini and Member of the Group Executive Committee. “As investment increases, the rise of more complex, autonomous AI systems signals a new era of generative AI that could impact the way companies operate. To propel their AI journeys forward, organisations should establish strong data foundations with clear processes to manage siloed data and enable data integration across functions. Trust, transparency, and accountability will continue to play a central role whilst embracing this next frontier of AI that has the potential to deliver significant value over time.”

AI chatbots are evolving to multi-agent systems[1] that are set to accelerate value creation

Almost three quarters of organisations (74%) agree that generative AI is helping them drive revenue and innovation. As AI technology progresses, it will transition from the role of supportive tool to that of independent agent with increased execution capability, allowing organisations to reimagine the way they do business and create greater value from their AI investments. This potential for value creation has contributed to the emergence of multi-agent systems – a quickly evolving technology with potential to drive increased innovation. The report indicates high levels of trust in AI agents for specific tasks such as generating professional emails, coding, and data analysis. However, it also reveals that leaders are conscious of the need to maintain this trust and ethical transparency in its development and deployment.

Increase in generative AI adoption across all domains includes the use of public tools

The acceleration of generative AI over the past 12 months is not limited to a business level, as recent technological developments have made public tools more accessible to non-experts. As a result, while there has been an increase in its adoption, only 3% of organisations have imposed a ban on the use of public generative AI tools in the workplace. Almost all organisations (97%) allow employees to use generative AI in some capacity, with over half enforcing specific guidelines for employees to follow.

Faced with the fast-paced uptake of generative AI, the report notes that organisations should proceed with a sense of responsibility. Clear guardrails to validate decisions made by multi-agent systems are essential to ensure transparency and accountability in operations and to mitigate the future risks that public tools may pose to their organisation.

To access the full report: https://www.capgemini.com/gb-en/insights/research-library/generative-ai-in-organisations-2024/

Methodology

The Capgemini Research Institute surveyed 1,100 executives employed at organisations with more than $1 billion in annual revenue across 14 countries including Australia, Canada, France, Germany, Italy, India, Japan, the Netherlands, Norway, Singapore, Spain, Sweden, the UK, and the US. Organisations operate across 11 sectors; nearly all (96%) of these organisations have started to explore generative AI. The global survey took place in May and June 2024. Executives surveyed are at director-level and above and represent diverse functions.

[1] Multi-agent systems are defined as technology designed to function independently, plan, reflect, pursue higher-level goals, and execute complex workflows with minimal or limited direct human oversight. Such systems exhibit characteristics traditionally found exclusively in human operators, including decision-making, planning, and adapting execution techniques based on inputs, predefined goals, and environmental considerations.