According to a report released Tuesday, API security breaches are costing businesses in the Asia Pacific (APAC) region a lot of money as they embrace artificial intelligence (AI) technology. The average predicted loss per incident is now more than $1 million.

According to an Akamai Technologies report, 81% of companies in the area had at least one API security breach in the last year.

The report said that the average cost per incident has increased significantly, from about $580,000 in last year’s analysis to over $1 million, indicating increasing cyber risks as AI acceptance picks up.

According to the report, the most frequent kind of security incident, cited by 43% of respondents, involved attacks utilizing APIs linked to AI technologies, such as apps, AI agents, and large language models (LLMs).

In addition, 93% of Indian firms and 90% of Singaporean businesses said they had experienced at least one API-related attack during the year, making them the most vulnerable to such assaults.

In the meanwhile, Japan had the greatest average financial impact from each breach at $1. 59 million, while Singapore came second with $1. 33 million for each occurrence.

According to the research, there is a significant lack of visibility in business security systems, with just 22% of survey participants claiming to have a comprehensive list of APIs and know which of them manage sensitive information.

In contrast, just 19% of organizations said they had completely integrated security testing into their API software development and deployment procedures, even though 72% of them said their focus on API security had increased in the previous year.

According to Reuben Koh, Director of Security Technology & Strategy at Akamai Technologies, Asia-Pacific & Japan, firms throughout the region are quickly implementing AI adoption, but the security infrastructures supporting those systems are still lacking.

He pointed out that APIs are increasingly becoming essential infrastructure for services powered by AI, and that insufficient supervision or transparency might result in service interruptions, increased recovery expenses, and erosion of public confidence.

The report warned that as AI services become more embedded into commercial operations, poor API visibility is becoming a compliance issue as well as a cybersecurity danger.