The Evolution of AI and Its Challenges and Opportunities in the Data Industry

AI has become a driving force in the data industry, where growing data volumes and complexity render traditional processing insufficient; AI offers more efficient analysis and higher decision value, and its role is expected to deepen as the technology matures.

The article first traces AI’s origins from Alan Turing’s 1950s thought experiment, through early pattern‑matching approaches, the rise of machine learning around 2000, and the recent explosion of data and compute that enabled neural networks, Transformers, and OpenAI’s GPT series.

It then discusses the application of AI—particularly generative models—in data analytics, highlighting that while 70% of data needs can be met with dashboards, the remaining 30% involve ad‑hoc queries where natural‑language interfaces face challenges such as low efficiency, data inconsistency, hallucinations, and lack of trust in generated SQL.

The article lists four key considerations for deploying GPT‑based data query assistants: (1) clearly define the scenarios and users for natural‑language data retrieval; (2) ensure the accuracy of returned data to build trust; (3) evaluate who benefits from efficiency gains and the business value; and (4) improve the usefulness of GPT‑generated content.

To address trust and consistency, the authors propose using a metric‑centered data platform as a guarantee layer, where standardized metric definitions and responsible owners resolve data‑caliber issues, allowing users to safely employ natural‑language queries for the remaining ad‑hoc needs.

Beyond end‑user queries, GPT can also assist in data processing, metadata generation, code explanation, and table retrieval; as Copilot‑style automation spreads, repetitive tasks will shift to machines, freeing humans for higher‑value work.