How Can We Get AI Right?

Have you ever heard about coded gaze? Neither had I, until I watched this amazing TED talk by Joy Buolamwini. And yes, I encourage you to watch it too. 8 minutes well worth your time.

According to TED, Joy Buolamwini was working with facial analysis software when she noticed a problem: the software didn’t detect her face — because the people who coded the algorithm hadn’t taught it to identify a broad range of skin tones and facial structures.

Although Artificial Intelligence (AI) is often assumed to be a rather technical domain, its relevance for business, democracy and politics becomes more and more evident as the technology integrates across various levels of our societies.

The impact of AI can be felt across sectors, from the labour market to the use of force. From controlling access to information on social platforms to algorithms controlling self driving cars. It affects everything from your daily information flow to the very institutions you have grown to take for granted.

Truth be told, AI and algorithmic systems already guide a vast array of decisions in both private and public sectors.

Sounds scary? Well, that’s only one side of the coin. It also brings immense opportunities, if we get it right.

The emergence of AI based technologies present great opportunities in areas as diverse as health, science, energy and sustainable development. But as with other revolutionary technologies before, many challenges arise with it. From strategic instability and volatility in the international arena and national power structures to coded bias and trust.

In short, AI based systems have the potential to influence and shape both the distribution of power, political practices and democracy.

Getting back to Joy Buolamwini and her story of coded bias. She realised later on that the facial analysis software she had been using for her project was actually widely used across the world. Yes, the one that could not detect her face because of coded bias — because the people who coded the algorithm had not taught it to identify a broad range of facial structures and skin tones. Imagine the impact of that.

Her research has also shown bias in softwares used by major companies as IBM and Microsoft. Again, one can only imagine the implications.

Today, an emerging number of actors are looking at these issues from various perspectives. Relating to both business, politics and civil society. Highlighting current problems and bringing attention to potential implications.

What is important to remember is that AI based tech is not only just about tech. Far from it. Its relevance for business, democracy and your everyday life makes it crucial to look at with critical eyes across various levels of our societies. And most importantly, involving actors from various backgrounds and sectors.