The race to adopt artificial intelligence (AI) is intensifying among businesses worldwide, with organizations leveraging this technology to enhance efficiency and streamline operations. However, the integration of AI is not as straightforward as it may seem, as it involves a complex web of components that must be monitored closely.
The Complexity of AI Integration
Akamai Technologies, a prominent content delivery network and cloud service provider based in the United States, has raised concerns regarding the intricate nature of AI deployment in companies. The reliance on various elements—such as models, application programming interfaces (APIs), databases, open source software, libraries, packages, and third-party services—means that one weak link can potentially disrupt the entire system.
Understanding 'Shadow AI Supply Chain'
During a recent virtual media briefing, Reuben Koh, Akamai's Security Technology and Strategy Director for the Asia Pacific and Japan region, introduced the concept of the 'Shadow AI Supply Chain.' This term refers to the use of AI components that are not always visible or monitored by organizations. Developers may not build every element of their products from scratch; instead, they often utilize open-source models and public repositories, which can introduce unforeseen risks.
Risks Associated with API Dependencies
The situation is further complicated by the modern AI’s heavy reliance on APIs. These interfaces serve as the connective tissue between AI models and various applications, databases, and external services. As AI takes actions within a business, APIs can also become vulnerable points of entry for cybercriminals. Recent data from Akamai highlights that this issue is already prevalent in the Asia Pacific region.
Rising Cyber Threats in the APAC Region
A 2026 API security study revealed that 43% of organizations surveyed in the APAC region experienced attacks involving APIs linked to AI technologies in the past year. Koh emphasized the importance for businesses to be aware of the APIs they are using, the systems they connect to, and the data they can access. Without this awareness, companies expose themselves to significant cybersecurity risks.
In conclusion, as AI continues to evolve and play a crucial role in business operations, organizations must prioritize monitoring and managing the multitude of components within their AI frameworks. Failure to do so could lead to severe vulnerabilities, impacting not only their systems but also their data security.
Source: https://tekno.kompas.com/read/2026/08/21/11340077/akamai--ai-di-perusahaan-punya-rantai-celah-satu-rusak-bisa-picu-banyak-masalah
