As IT resource demands soar, 97% of enterprise IT leaders in the Asia-Pacific (APAC) region have either implemented or plan to implement AI agents within the next two years. This shift aims to ease workloads and improve productivity, but many companies face significant hurdles in achieving full integration. A new report reveals that 95% of APAC organisations struggle with integrating data across systems, limiting the potential of AI agents to create a seamless digital workforce.
On average, only 27% of an organisation’s applications are connected, reducing the accuracy and effectiveness of AI agents. Integration is crucial because these agents rely on connected data from systems like CRM, ERP, and HCM, as well as sources such as emails, PDFs, and collaboration tools like Slack. By bridging these gaps, organisations can enable AI-powered agents to work autonomously and manage both simple and complex tasks across various business functions.
Demand for IT solutions drives AI adoption
The rise in IT resource demand is opening doors for AI agents to increase efficiency and cut costs. According to MuleSoft’s 10th annual Connectivity Benchmark Report, IT teams already face challenges maintaining existing systems while tackling new AI-based projects.
In 2024, 27% of IT projects in APAC were delayed, compared to 25% the previous year. The pressure is mounting, with 85% of IT leaders expecting workloads to increase. On average, IT leaders foresee a 16% growth in project volume between 2024 and 2025. To meet these demands, APAC enterprises spent an average of US$11.2 million on IT staffing in 2024, a rise from US$10.2 million in 2023.
AI agents are seen as a solution to this growing demand, with 93% of IT leaders believing that AI will boost developer productivity within the next three years. Those already using AI agents are even more optimistic, with 98% expressing confidence in their benefits. Currently, around 39% of IT teams’ time is spent on building and testing custom integrations between systems—a task that can be streamlined through automation and integration solutions.
Complex infrastructures limit AI effectiveness
The increasing number of applications and AI models used by businesses is creating data silos, hindering AI agents from operating effectively. The report highlights that 93% of APAC IT leaders say data silos are causing business challenges. The issue is more pronounced among organisations using AI agents (98%) compared to those that don’t (89%).
Today, APAC organisations use an average of 912 applications. Those with AI agents have even more, averaging 1,130 applications—47% higher than organisations without agents. The number of AI models has also surged, with companies using 19 models on average in 2025, up from eight in 2024. Organisations leveraging AI agents rely on 24 models on average, while those without agents use about 15.
APIs and integration solutions improve AI performance
To overcome these challenges, aligning integration strategies with AI deployment is critical. Organisations can unlock their data’s full potential by adopting APIs, which allow AI agents to access essential business-specific data and perform tasks efficiently. Among APAC businesses using AI agents, those leveraging APIs report key improvements, including:
- Enhanced IT infrastructure (50%)
- Improved data sharing across teams (48%)
- Automated workflows (46%)
On average, 48% of an organisation’s internal software assets and components are reusable, creating an opportunity to build AI-powered solutions without starting from scratch. APIs also offer broader business benefits, such as increased productivity (52%) and faster responses to business needs (49%). APAC IT leaders estimate that 38% of their company’s revenue is linked to APIs, rising to 45% for organisations using AI agents.
Expert perspectives on integration and AI
Andrew Comstock, SVP and GM of MuleSoft, stressed the importance of integration, saying, “AI agents are set to transform enterprises with limitless digital labour. We see forecasts of new levels of efficiency and productivity with the unified human and agent workforce. However, to unlock this new capacity, integration and APIs are critical to building an agent-ready foundation.”
Beena Ammanath, Global Deloitte AI Institute Leader, added, “In the agentic era, IT leaders are seeking ways to meet the increased demand for efficiency and productivity. Leading organisations must establish a proactive integration strategy for unifying the entire IT estate, encompassing apps and systems, automation, and APIs, all of which are vital for driving revenue growth and reducing operational costs.”
Maximilian Schroeck, Global Chief Commercial Officer for Salesforce at Deloitte Global, commented, “This report provides invaluable guidance for businesses navigating the complexities of digital transformation and the underlying need for effective data integration. By leveraging these insights, we empower organisations to unlock the full potential of integration—connecting systems, data, and processes to drive innovation, enhance customer experiences, and achieve sustainable growth.”
