We need more bias in artificial intelligence

What makes one vision more desirable than another is not its neutrality, but whether it can better serve one’s goals in the context of where those goals are being pursued.

By: Date: April 21, 2021 Topic: European Macroeconomics & Governance

This opinion piece was originally published Il Sole 24 Ore.

logo of Il Sole 24 Ore Italian newspaper

The Muller-Lyer optical illusion consists of two lines of equal length that differ only in the direction of arrowheads at either end. Yet, to most observers, the line with arrowheads pointing outwards looks longer than the other. If you grew up in and among buildings with straight walls and 90 degree angles, you have learned to perceive lines according to geometric patterns. Your view of the Muller-Lyer lines is biased.

Artificial intelligence developers sometimes fall into similar traps. They program life-changing applications, into which they project biases. Algorithms have led judges to be harsher on Black offenders when assessing the likelihood that defendants will commit the same crime again. Machine-learning applications favour male over female job applicants (Amazon scrapped its recruiting tool after the system learned to attach a lower score to applications that mentioned the word ‘woman’). Automatic translation programs replicate gender stereotypes. For example translating from a language without gendered pronouns (such as Finnish) into English, the algorithm could suggest using ‘he’ when the action is ‘to work’ and ‘she’ when the action is ‘to wash the laundry’.

We should be very concerned about bias embedded in artificial intelligence. Efforts by public authorities to curb it, such as the regulation being proposed by the European Commission in ‘the Artificial Intelligence Act’, are welcome.

Often, however, biases are not only embedded in the design of the algorithm. They are also external to it, originating in societal biases. Amazon’s recruiting tool inherited the bias from a dataset covering a decade during which most job applications came from men (a symptom of the strong asymmetry in gender power in the technological sector). Similarly, automated translation applications learn gender stereotypes from the thousands of books used to train them. Discrimination against women and minorities is well reflected in literature.

No matter how objective we try to be, the mere decision to adopt artificial intelligence solutions has profound implications. That decision is inherently subjective and thus comes with some political responsibility, which goes beyond simply regulating the use of artificial intelligence. Algorithms learn to be as discriminatory as the society they observe. They then suggest decisions that are inherently discriminatory, and thus contribute to exacerbate discrimination within society. Policy may break this vicious circle.

If public policy aims to improve decision-making and build a more inclusive society, it should deal explicitly with the question of the role of artificial intelligence in achieving the end goal. If artificial intelligence amplifies society’s biases, policy may well need to intervene, either prohibiting its use or embedding counterbalancing biases. For example, algorithms that automatically rank subjective content in online chats could be compelled to attach lower weights to discriminatory comments. This, in effect, would distort the sentiments of a community: perhaps in the collective image of a community populated by men, women are not associated with intellectual work. But the algorithm would then yield a representation of the world closer to what we would like it to be. In medical research, desirable biases could be used to correct gender imbalances. Coronary heart disease is a leading cause of death for women, but men are overrepresented in clinical trials: artificial intelligence could favour women’s enrolment over that of men.

This does not mean that politicians should systematically interfere with technology markets, micromanaging technology development and adoption. But an overall political vision is needed to set the direction of travel, if the aim is to live in a better world.

We often already call for the introduction of desirable biases through affirmative action. Gender quotas address discrimination against women in the selection for positions of power. Quotas do not however simply correct bias. They are also a political statement: gender equality is a tool to change the system structurally. Male-driven decision making in companies or public institutions could indefinitely perpetuate itself, with those in charge continuing to select those who match their male-oriented vision of the world. Imposing quotas is tantamount introducing a bias against that; it means rejecting one way of doing things and instead supporting a different vision that aims to correct historic marginalisation.

Similarly, the discussion on how to improve the use of artificial intelligence in Europe should not be separated from its structural implications.

In the 1960s, anthropologists realised that members of Zulu tribes in South Africa did not fall for the Muller-Lyer illusion. Unlike their peers from Western societies, they saw immediately the lines were of the same length. Their interpretation of the information provided was different. Zulus live in rounded huts in an environment where the sharp angles of European buildings are absent. Their geometric vision is different. Of course, a Zulu might find herself less at ease estimating distances in a European city.

Ultimately, what makes one vision more desirable than another is not its neutrality, but whether it can better serve one’s goals in the context of where those goals are being pursued.

This was produced within the project ‘Future of Work and Inclusive Growth in Europe’, with the financial support of the Mastercard Center for Inclusive Growth.

Republishing and referencing

Bruegel considers itself a public good and takes no institutional standpoint.

Due to copyright agreements we ask that you kindly email request to republish opinions that have appeared in print to [email protected].

Read article Download PDF More on this topic More by this author

Working Paper

Research and innovation policies and productivity growth

Can research and innovation policies power growth? The answer currently can only be a timid yes. Too little is known of what drives the actual effects of R&I policies.

By: Reinhilde Veugelers Topic: Innovation & Competition Policy Date: May 10, 2021
Read article More on this topic More by this author

Blog Post

Algorithmic management is the past, not the future of work

Algorithmic management is the twenty-first century’s scientific management. Job quality measures should be included explicitly in health and safety risk assessments for workplace artificial-intelligence systems.

By: Laura Nurski Topic: Innovation & Competition Policy Date: May 6, 2021
Read about event More on this topic

Past Event

Past Event

Covid-19 and the geopolitics of the Balkans

How have China, Russia, Turkey and others stepped up their activities in the Balkans at a time when the prospect of enlargement is diminished?

Speakers: Michael Leigh, Pierre Mirel, Aleksandra Tomanić, Justyna Szczudlik and Catherine Wendt Topic: Global Economics & Governance Date: April 29, 2021
Read article More on this topic

Blog Post

Setting Europe’s economic recovery in motion: a first look at national plans

Plans for spending European Union recovery funds submitted by the four largest EU countries reflect rather different priorities. So far, only Italy is interested in borrowing from the EU.

By: Zsolt Darvas and Simone Tagliapietra Topic: European Macroeconomics & Governance Date: April 29, 2021
Read article More on this topic


The ECB needs political guidance on secondary objectives

While EU Treaties clearly stipulate that the ECB “shall support the general objectives of the European Union”, it is not appropriate to simply stand by, wishing that the ECB will use its discretionary power to act on them. Political institutions of the EU should prioritise the secondary goals to legitimise the ECB’s action.

By: Pervenche Béres, Grégory Claeys, Nik de Boer, Panicos O. Demetriades, Sebastian Diessner, Stanislas Jourdan, Jens van ‘t Klooster and Vivien Schmidt Topic: European Macroeconomics & Governance Date: April 22, 2021
Read about event More on this topic

Past Event

Past Event

AI regulation at the service of industrial policy?

What role should the EU play in the regulation of AI?

Speakers: Julia Anderson, Joanna Bryson, Annika Linck and Martin Ulbrich Topic: Innovation & Competition Policy Date: April 22, 2021
Read article

Blog Post

Urgent reform of the EU resolution framework is needed

In this blog, the authors argue that two aspects of the European resolution framework are particularly in need of reform – the bail-in regime and the resolution mechanism for cross-border banks – and propose a reform of both.

By: Mathias Dewatripont, Lucrezia Reichlin and André Sapir Topic: European Macroeconomics & Governance, Finance & Financial Regulation Date: April 16, 2021
Read article More on this topic

Blog Post

The impact of COVID-19 on artificial intelligence in banking

COVID-19 has not dampened the appetite of European banks for machine learning and data science, but may in the short term have limited their artificial-intelligence investment capacity.

By: Julia Anderson, David Bholat, Mohammed Gharbawi and Oliver Thew Topic: Finance & Financial Regulation Date: April 15, 2021
Read article More on this topic More by this author


More Europe or less Europe?

Europe is often a ship with multiple captains. The boat moves forward in calm seas, but when the slightest wind puts it off course, it is not easy to steer that boat. It is not so much a question of more Europe rather than less, but of achieving ‘one Europe’. A ‘more-or-less Europe’ is an invitation to go nowhere.

By: Maria Demertzis Topic: European Macroeconomics & Governance Date: April 14, 2021
Read article More on this topic More by this author



Keeping momentum on good governance

Transparency, human rights and good governance: a conversation with Katalin Cseh MEP

By: The Sound of Economics Topic: European Macroeconomics & Governance Date: March 17, 2021
Read about event More on this topic

Past Event

Past Event

Protecting workers in the platform economy

How can we protect platform workers while preserving the opportunities and benefits that are generated by the sharing economy?

Speakers: Payal Dalal, Mario Mariniello, Diane Mulcahy, Ana Carla Pereira, Jacob Rudbäck and Shamina Singh Topic: Innovation & Competition Policy Date: February 24, 2021
Read article More by this author


Will COVID accelerate productivity growth?

The COVID-19 pandemic has prompted an increasing number of rich-country firms to reduce their reliance on global supply chains and invest more in robots at home. But it is probably too soon to tell whether this switch will increase productivity growth in advanced economies.

By: Dalia Marin Topic: European Macroeconomics & Governance, Global Economics & Governance Date: February 10, 2021
Load more posts