Intel announced on the 10th that it had commissioned IDC to conduct the '2024 IDC Asia/Pacific AI Maturity Study'. The survey was conducted on AI maturity in eight countries in the Asia-Pacific region, including Korea (Australia, India, Indonesia, Japan, Korea, Malaysia, Singapore, and Taiwan).

According to this research survey, the overall AI maturity of the Asia-Pacific region is at an intermediate level, and Korea was evaluated as above average in all three aspects: business, government, and socio-economics.

AI spending in the Asia Pacific region is expected to grow at a CAGR of 28.9% through 2027, reaching $90.7 billion (KRW 124 trillion). IDC states that "with the rapid rise of generative AI (GenAI), many organizations have begun to adopt AI for internal AI initiatives aimed at improving productivity and reducing costs starting in 2023."

What’s interesting about the report is that despite the generative AI craze, only 19% of AI budgets in the Asia-Pacific region are allocated to generative AI. With the remaining 81% allocated to predictive and interpretive AI, IDC added that AI should be viewed as a continuum, saying, “Predictive, interpretive, and generative AI are all important and should not be considered separate areas.”

AI at the edge, or on-device AI, is gaining much attention and is rapidly emerging as a keyword this year. While it is important to implement AI in real time, there are many cases where it is not properly implemented due to latency issues related to centralized infrastructure. The transition to edge computing is expected to enable AI to be implemented everywhere.

More than 75% of APAC organizations surveyed expect to spend more on edge by 2024, with nearly 50% estimating that edge will account for more than 16% of overall IT spending.

As the paradigm shifts to the hybrid AI era, the need for technology infrastructure that can distribute and manage data and computing from data centers to clouds and from clients to the edge is increasing to meet the diverse requirements of model learning and inference.

IDC currently evaluates the AI maturity level of each country in the Asia Pacific region as follows: Indonesia and Malaysia at Level 1 (Exploration), India and Taiwan at Level 2 (AI Practice), Australia, Japan, and Korea at Level 3 (AI Innovation), and Singapore at Level 4 (Leader).

Korea is currently rated as an AI Innovator in the 2024 IDC Asia Pacific AI Maturity Study, highlighting that it is in the stage of planning and managing AI initiatives with well-established technology infrastructure and data management strategies, and where new use cases, especially industrial use cases, are emerging.

Korea's investment in AI infrastructure provisioning is estimated at $386.32 million, supporting extensive AI development. These large investments are being used to improve computing resources such as high-end servers and dedicated AI processing units. South Korea appears to be prioritizing the use of AI to transform customer service, with a particular focus on supplementing customer service agents.

Looking at the industrial sector, it showed a tendency to focus on: △large-scale AI investment to support precision manufacturing automation in the semiconductor sector and use of ML for quality control; △use of high-spec infrastructure using AI for complex financial analysis in the BFSI sector; and △focus on improving automation efficiency and, in particular, digital assistants and service personalization in other sectors such as distribution, communications, healthcare, and education.

Moreover, contrary to the common belief that GPUs are required for all AI workloads, in reality, general-purpose CPUs are often used as a more efficient alternative for performing some AI-based tasks, which can contribute to sustainability by reducing costs and energy consumption.

Intel is attempting to provide cost advantages for dedicated AI accelerators such as Gaudi 3, thereby providing an alternative to existing GPUs. However, experts predict that it will not simply be a battle between CPUs and GPUs, and that the adoption of solutions for each demand application will differ.

IDC suggested the following recommendations for increasing the AI maturity of the market: △Preparing for hybrid AI, △Building a flexible and scalable technology base, △Adjusting AI regulations and ethics at the government level, and △Increasing AI understanding among experts and non-experts.