About 19% of AI use cases deliver business value, says Infosys report

The research also showed a strong link between AI success and the changes a business makes to its operating model and data structure

About 19 per cent of artificial intelligence (AI) use cases deliver on all their business objectives while about 32 per cent show promise by partially meeting their objectives, a research by Infosys showed.

 

The research, surveyed across 3,240 companies and 132 different AI business use cases, also showed a strong link between AI success and the changes a business makes to its operating model and data structure.

 

Enterprises across business verticals have seen less-than-expected adoption of AI and GenAI as they remain wary of the costs, return on investments and reluctance among its employees to use it.

 

However, the arrival of China’s Deepseek has raised expectations of faster adoption due to its cheaper language model.

 

“As AI costs decline in the future, the research data indicates these transformational use cases will rapidly begin to deliver more effective business outcomes. We will see an acceleration of AI use cases that achieve viability for businesses resulting in the mass expansion of AI agents across an average enterprise in 2025,” the report titled, AI Business Value Radar, mentioned.

 

IT, operations, and facilities are the most pursued AI use case category, with 38% of respondents implementing it and showing more viability. This is followed by cybersecurity and resilience (30 per cent), software development (30 per cent), marketing (26 per cent) and customer service (24 per cent). Use cases in these categories are 10-15 per cent more likely to succeed.

 

The report also pointed out various use cases that do not need huge transformation programs or costs for companies to implement. Those include IT operations, software development, chatbots, financial reporting and cash flow forecasting, fraud detection and risk analytics, marketing asset creation and supply chain optimisation.

 

“These are good bets for companies looking for low risk, low effort and potentially lower cost AI wins. This is because they require much less change to the organisation and data structure and they are more likely to be successful with relatively low levels of spending,” the report added.

 

Professional services, life sciences, high tech, telecommunications, and insurance industries are more likely to achieve success from AI implementation while financial services may not fare better due to regulatory and data modernisation challenges.

 

However, AI is not benefiting all industries equally with travel and hospitality, manufacturing, retail, and the public sector struggling to achieve consistent success.

 

The most important challenge organisations face is change management as engaging the workforce is the most important driver of successful AI outcomes.

 

The report highlighted that only 16 per cent of the surveyed firms have implemented effective change management and employee training for AI. Companies that have taken initial steps to address AI can nearly double their likelihood of success with AI deployments by fully investing in workforce AI readiness.

 

“Those companies that leave their workforce out of the equation risk are missing out on the promised benefits of enterprise AI era,” it said.