How Can CIOs Course-Correct AI for Inherited Human Biases?
InformationWeek, Friday, October 23rd, 2026
AI inherits bias from the data and decisions people write down, and CIOs explain how ownership and oversight can counter it.
AI may be neutral in theory, but it inherits bias from developers, users and the records it learns from.
VarOps CTO Ran Aroussi notes that AI reasons over what got written down, which is often skewed, so leadership decisions misremembered in chat threads can lead the system to answer confidently as if it saw the whole company.
Infoblox CIO Justin Kappers says every consequential AI-assisted decision at his company has a named business owner, with oversight scaling with autonomy: a document summarizer needs light review, but an agent that influences financial transactions needs human checks.
The piece urges CIOs to make sure teams understand how their own prejudices shape AI outputs.