AI Is Getting Excel Wrong – What Businesses Need to Know

Durban: As artificial intelligence tools like ChatGPT and Microsoft Copilot become more widely used in the workplace, many organisations are seeing a significant increase in productivity. However, alongside these benefits, a growing risk is emerging — incorrect Excel outputs that appear accurate.

In many corporate environments, Excel remains a core tool for financial reporting, HR analysis, and operational decision-making. AI tools are now being used to summarise data, generate formulas, and automate analysis.

While this can save time, it can also introduce errors that are not immediately obvious.

The challenge is not the AI itself — it is how it is being used.

AI tools generate responses based on patterns, not actual data validation. This means outputs can look structured and professional while still being incorrect. Common issues include missing data due to filters, incorrect assumptions about data ranges, and summaries that do not reflect the full dataset.

In Excel, these risks often manifest as common formula errors, including incorrect references, mismatched data types, and calculation issues that go unnoticed without proper validation.

This creates a new requirement for organisations: employees must not only know how to use AI tools, but also how to verify and validate the results.

As AI adoption continues to grow, businesses should focus on three key areas:

Understanding common Excel errors
Implementing validation processes
Training employees to use AI responsibly

AI can significantly enhance productivity — but only when used with the right controls in place.

For a practical breakdown of common Excel errors and how to address them, read:

https://collegeafricagroup.com/types-of-errors-in-excel/

Comment from College Africa Group

“AI produces language — not truth. In our training sessions, we’re seeing more teams rely on outputs that look correct but haven’t been validated. The real risk is not the tool itself, but how quickly results are accepted without checking the underlying data.”
— Arnold Muscat, Director, College Africa Group