AI and the Reality of Data: Garbage In, Garbage Out

Artificial intelligence, or AI, has now entered various aspects of our lives. But when using AI, we often forget a fundamental truth—the better you train AI, the better the results will be. In other words, if the data is incorrect, incomplete, or lacks context, the output from AI will also be wrong. In the words of technology experts, “Garbage in, garbage out.”

In a recent discussion, one speaker said, “Our challenge with AI is that when we look for general information, AI provides a lot of data. But much of that data is not relevant to Singapore or our own context. As a result, we have to rely on our internal databases.” He further added, “Unfortunately, if your database contains garbage, then you’ll get garbage from AI as well. That’s why we are now working on improving the quality of that information.”

This statement highlights an important reality—no matter how advanced AI becomes, its effectiveness depends on the data it is trained with. AI models built on Western data often don’t apply to the socio-administrative realities of Asia or Singapore. For instance, in areas like healthcare, education, or public services, if local policies, culture, language, and demographic differences are ignored, AI-driven decisions can often be wrong.

The speaker also said, “We are now working on many projects, but the question is—how do we use AI to make better use of the information we already have? At the moment, we are somewhat empty-handed, which makes the situation a bit chaotic.” These remarks are not just about technology management issues; they offer a deeper lesson—the foundation of a data-driven future lies in the quality of the data.

In this context, organizations must now consider how up-to-date, reliable, and relevant their internal databases are. Artificial intelligence will work effectively only when the data it is fed aligns with reality. Otherwise, no matter how advanced AI is, the results will be just as confusing.

In today’s technology-driven world, “Data is the new oil”—but just as unrefined oil can damage an engine, unrefined or incorrect data can cause chaos in AI’s outcomes. So the most important investment for the future lies not only in artificial intelligence, but in the quality of the data that feeds that intelligence.

This reality reminds us that no matter how advanced technology becomes, it can never be separated from the human context. Because accurate information is the true lifeblood of artificial intelligence.