One important area where design choices can help mitigate risks associated with new technologies such as AI is privacy, which is likely to become increasingly salient for organisations using data to drive growth because they will inevitably require access not just to large pools of data but also sensitive personal information such as income level or health-statusised profiles based on their online behaviours.

By contrast, social media platforms collect reams of highly-specific customer information through interactions with their users while sharing only limited amounts of it with advertisers. To curb these practices, regulators have begun requiring companies that collect user data – including Facebook – to disclose what they do with it and although most companies share only aggregated versions of Personally Identifiable Information (PII), there have been instances where businesses were found collecting PII without consumers knowing it was happening. One well-publicised example involved Amazon employees listening into some Alexa users’ conversations without permission. These incidents underscore how important it is for executives responsible for developing new products built around big data and machine learning systems to consider how user privacy will be protected early enough during development processes so they can avoid problems later when products are already deployed. Transparency is key when collecting sensitive information.