A new study commissioned by tech titan IBM has revealed that Asia-Pacific enterprises are moving beyond AI experimentation to maximize the impact of their AI investments.
More than half (54%) now expect AI to deliver longer-term benefits for their business in areas such as innovation or revenue generation, according to the study titled “APAC AI Outlook 2025”.
The game changer lies in developing cost-effective AI solutions, with the flexibility to use custom-built open-source models and seamless integrations between multiple vendors, the report said.
The study noted that the pursuit of short-term wins during the initial phase of generative AI projects has given way to a more nuanced understanding of AI’s potential.
The focus is also shifting from low-risk, non-core use cases, to deploying Gen AI in core business functions for competitive advantage and improved ROI.
According to the report, conducted by Ecosystm on behalf of IBM, nearly 60% of surveyed organizations across the Asia-Pacific region anticipate realizing the benefits of their AI investments within two to five years. Only 11% expect returns within the next two years.
In 2025, the primary focus of AI investments for Asia-Pacific organizations will center on enhancing customer experience (21%), back-office business process automation (18%) and sales automation and customer lifecycle management (16%).
To realize these goals, the report said organizations must address key challenges, including data complexity (39%), high cost of implementation and solution (36%) and limited number of use cases defined (35%).
The five strategic trends shaping the region’s AI future identified in the report includes:
1. AI-led revenue generation takes center-stage: Organizations will adopt a “Strategic AI” approach in 2025, prioritizing projects based on feasibility and business impact. This reflects maturing recognition that early wins to build trust and organizational buy-in must be balanced with longer-term AI strategies. The challenge is how to scale AI through use cases that maximize revenue opportunities and ROI.
2. Smaller specialized open-source models emerge as a powerful alternative for many AI applications: Purpose-built models will be in demand, including those designed for local languages, nuanced regional contexts and simpler computational tasks. The “Rightsizing AI” models require significantly less training data and generate a smaller carbon footprint than the large language models that have so far dominated AI discussions.
3. Enterprises embrace new tools for visibility, governance and seamless AI integration: Asia-Pacific organizations will increasingly leverage open-source AI models to drive innovation and efficiency. The “Unified AI” with robust orchestration tools will streamline the management around these solutions, offering flexibility, cost-effectiveness, improved security and seamless integrations between different vendors.
4. AI agents redefine the future of work: Enterprises will increasingly devise agentic workflows, powered by AI agents, to autonomously execute tasks, collaborate with human workers and drive value across the business. The “Agentic AI”, combining AI with automation, has the potential to achieve significant gains in operational efficiency, customer experience and decision-making. However, organizations need to establish internal guardrails, and regularly evaluate underlying models to ensure ethical and responsible use.
5. Human-centered innovation drives the next phase of AI: While productivity tools have been a major focus of AI adoption, the future lies in leveraging AI to enhance human experiences and capabilities. “The Human-Centric AI approach” will become a powerful tool for employees to augment their roles, automate routine tasks, and unlock new opportunities for creativity and innovation. By prioritizing the empathetic design of AI solutions, organizations can also foster stronger customer relationships and brand loyalty.
For the Philippines, the IBM study cited the AI experience of two local organizations – tech solutions provider eBiZolution and government-owned Philippine Rice Research Institute (PhilRice).
In the case of eBiZolution, which integrates AI across its business units to elevate its solution portfolio and value-added services, its key initiatives that include AI are:
- Exposure Management System for unified visibility and continuous data analysis of the attack surface for proactive incident detection and response.
- Intelligent Video Analytics for real-time actionable intelligence, including license plate recognition, facial recognition, and behavioral analysis.
- Talent Optimization Platform, currently being developed on watsonx, is an internal talent management tool that targets to intelligently qualify personnel for private and government tender requirements to streamline workforce allocation, which will enhance efficiency and maximize success in competitive bids.
eBiZolution founder and CEO Nathaniel Marquez said the company’s AI-driven solutions foster innovation and enhance security in its solution portfolio in the following manner:
- The Exposure Management System ensures compliance with frameworks like NIST and ISO/IEC while safeguarding critical digital assets and sensitive information.
- Intelligent Video Analytics addresses ethical concerns through responsible monitoring practices, delivering precise insights while protecting individual rights.
However, there are challenges that come with AI deployment such as privacy concerns for video analytics. In this case, the company is putting up robust safeguards such as selective monitoring and centralized AI processing to protect sensitive data.
But the company said data readiness remains a barrier as many private and public organizations lack robust infrastructure, limiting AI’s immediate potential.
In the future, eBiZolution plans to explore the automation of its infrastructure management workflows to optimize the time-to-service and efficiency of its managed services.
Over the next two years, the company anticipates increased AI adoption among clients by delivering improved reporting, advanced analytics, and streamlined operations. However, human oversight will remain integral to ensuring accountability and precision in critical decision-making.
In the case of PhilRice, an agency under the Department of Agriculture (DA) tasked with developing high-yielding, cost-reducing technologies to ensure rice sufficiency in the Philippines, it is integrating AI into key initiatives, including:
- Pinoy Farmers’ Text Center: An SMS-based chatbot providing real-time support for rice-related queries, such as identifying high-yield varieties.
- RiceLytics: A data analytics and visualization platform offering insights on rice production, self-sufficiency, and consumption. Plans include integrating satellite imagery and pest information, leveraging IBM Environmental Intelligence to enhance climate risk management.
- Philippine Rice Information System (PRiSM): Southeast Asia’s first rice monitoring system utilizing satellite imagery. PhilRice aims to incorporate AI and machine learning for improved planning, decision-making, and disaster preparedness.
• Digital Tools:
- PalayCheck App: provides farmers with actionable insights on best practices in crop management, optimizing operations, and ensuring sustainability. Farmers may access cropping calendar where they can be advised of the best time to start land preparation up to harvest time. Appropriate crop management guidelines can also be accessed.
- RCMAS (Rice Crop Manager Advisory Service): digital service agriculture service that aims to provide farmers with precise information to enhance productivity and profitability combined with Integrated Crop Management (ICM) developed by IRRI and PhilRice.
- PRIME (Pest Risk Identification and Management): provides timely information on pest outbreaks to better plan program interventions and field management strategies to avoid or mitigate crop losses due to pests and diseases.
Though still in its early stages, PhilRice expects AI to enhance decision-making, traceability, planning, risk management, and climate resiliency, though it remains in the early stages of adoption.
As a public entity, various local and international organizations from the public and private sector are seeking to use PhilRice’s historical data to train AI models and develop solutions, which raises concerns in sharing and managing the data due to privacy, ownership and provenance, as well as cost of subscribing to the solutions that would be developed. Data readiness remains a barrier but PhilRice is planning adopt a hybrid cloud approach.
PhilRice is also looking to automate HR, inventory, and financial systems while investing in workforce upskilling for digital transformation. It envisions its AI-driven analytics being adopted across the DA to expand beyond rice to other crops.
The DA also plans to leverage AI for improved productivity and efficiency across sectors, implementing data lakes, enhancing knowledge management, and advancing Industry 4.0 transitions with cloud-based and IoT solutions.
