TiDB warns outdated data could drive costly AI transaction errors

Traders Union
All News
Business & Wealth
Companies
TiDB warns of AI data errors

Highlights

TiDB's Predee Kajonpai warns that outdated or unsynchronized data increases risk of costly errors as AI agents handle financial transactions autonomously.
He cites that up to 95% of enterprise AI initiatives fail to reach full production and 40% of AI projects could be canceled by 2027.
Kajonpai advocates unified data platforms for real-time transactions and analytics to minimize repeated payments and improve AI reliability in financial operations.

Enterprise AI risks in financial operations

As reported by BusinessWorld Online, TiDB Senior Solution Architect Predee Kajonpai says businesses using AI agents for financial transactions face growing risks when underlying data is outdated or spread across unsynchronized systems. Speaking at the AI Vision Forum Manila on Wednesday, he says these problems become more serious as companies move from conversational AI tools to autonomous systems that can execute business actions on their own.

Kajonpai says an incorrect chatbot response can simply be retried, but mistakes in loan approvals, payments, or refunds carry direct financial consequences. He gives a hypothetical example in which an AI agent issues a customer refund, then repeats the payment after a system interruption because it fails to detect that the first transaction has already been completed.

He says businesses need to resolve these data management gaps before deploying AI agents in operations that depend on accurate and current information. Kajonpai also cites estimates that up to 95% of enterprise AI initiatives fail to reach full production, and notes that 40% of AI projects could be canceled by 2027.

Unified platforms seen as a way to reduce errors

Kajonpai says outdated and fragmented information systems are among the technical barriers affecting the rollout of autonomous AI applications. He advocates unified data platforms that handle both transactions and analytics in real time, reducing the need to copy data across separate systems.

Such platforms can help organizations maintain consistent transaction records and lower the chance that AI agents repeat completed actions, he says. TiDB provides database technology designed to support transaction processing, real-time analytics, and AI-related applications, positioning this approach as a way to improve reliability in automated business operations.

Gartner previously projects that more than 40% of agentic AI projects will be canceled by the end of 2027 because of rising costs, unclear business value, or inadequate risk controls. That industry outlook reinforces concerns that technical readiness, especially around live and synchronized data, remains central to whether AI systems can move safely into production.

In our earlier article on Cohere’s planned acquisition of AI cybersecurity startup AISLE alongside its tie-up with Aleph Alpha, we outlined how the company is strengthening enterprise security as AI systems become more capable. We noted that AISLE’s vulnerability-detection technology and research track record are aimed at helping organizations spot and fix software flaws, supporting safer deployment for enterprise and public-sector use cases.

This material may contain third-party opinions, none of the data and information on this webpage constitutes investment advice according to our Disclaimer. While we adhere to strict Editorial Integrity, this post may contain references to products from our partners.