Why the world needs more Women in Tech

This time last year, to mark International Women’s Day, the CBI had its first Women in Technology network meeting and I’m delighted to say that the group is now going from strength to strength.

We set the group up because technology is shaping the future, and if that’s going to be a future that works for women, then it’s vital that women are shaping technology. The excellent Caroline Criado Perez has written extensively about how the world is designed around a “one-size-fits-men” approach, to the detriment of women and girls.

I have my own personal experience of that. My Dad is a helicopter pilot, and – because of his love of aviation – my middle name is Amy, after Amy Johnson, a pioneering aviator who set many long-distance flying records during the 1930s. Exactly the kind of woman we should be celebrating on International Women’s Day!

The indomitable Amy Johnson

My Dad was keen to share his love of flying with my sisters and me. I have fond memories families’ days and the highlight of seeing the helicopters when I was a little girl.

A person sitting on a chair

Description automatically generated
Me as a little girl exploring a helicopter

But one memory is particularly deeply ingrained. It’s the moment when I knew for sure I wasn’t going to take to the skies. My dad was telling me how long some of their flights sometimes were, and he pulled out this tube and asked me what I thought it was. I had no idea. Cue awkward moments of silence while I tried to figure out what this thing could possibly be… turned out it was the toilet. Well let’s face it, that isn’t really going to work if you’re a girl.

Now, I’ll be honest, I was probably never going to be a pilot anyway, I still haven’t learnt to drive yet (I’d hoped cars would all be driverless by now) but it’s a reminder that if things aren’t designed by women, they probably won’t work for women.

Technology needs to be designed, made and delivered by a diverse group people

Technologies like AI are reshaping the way the world works in all sorts of ways. If that’s going to have positive social outcomes, it’s vital that businesses who are using AI do so in a way that ensures women and minority groups are represented.

If algorithms are trained on data that is incomplete or contains historic prejudice against particular groups, it will learn those unfair biases and incorporate them into its decisions, entrenching existing unfairness and barriers. Taking action to reflect diversity and inclusion in teams and data will lead to social benefits and makes good business sense too. Businesses who prioritise fairness and inclusion are more likely to create algorithms that make better decisions, giving them the competitive edge.

At the CBI we’re keen to help our members adopt technology well, in a way that works for everyone.

Business leaders can take action by actively championing diversity, and making sure the data they use is representative

Actively champion diversity:

  • Develop stretching targets and metrics to work towards fairness, and continue to collect data on, monitor, and adjust them.
  • Leaders should champion diversity and make it a shared priority across their business.
  • Question pre-existing practices: For example, questioning interview panels or candidate shortlists that have the same type of person will lead to more diverse teams, which have the range of perspectives that can design more inclusive technology.

Make sure your data is representative:

  • Use available technical tools to test and monitor algorithms and datasets such as IBM’s Fairness 360 Kit, an open-source toolkit containing metrics to check for unwanted bias; and Google’s What-If Tool, which lets users try five different mathematical types of types of fairness, as well as allowing them to analyse the performance of their AI models with new data or datasets.
  • When data is incomplete or contains historical unfair bias, take steps to make it more inclusive, and collect new data with AI in mind. Data that is incomplete or contains unfair bias can be cleaned. Businesses are also taking steps like collecting more representative data to improve the inclusiveness of their datasets. Businesses should also think about potential AI use cases you may have in a year or two to help define the fresh data you would like to collect.

CBI members can read more about putting AI ethics into practice in our member guide.

The CBI’s Women in Technology Group exists to help ensure women’s voices are heard

A group of people standing in front of a crowd posing for the camera

Description automatically generated
The members of the CBI’s Women in Tech group at our Annual Conference, 2019

Not only do diverse voices need to be heard in the development of technology, but in they also need to be heard in the development of the policy that sits around that technology. Our Women in Technology group exists to make sure the brilliant women in our membership have a specific chance to feed into our policy work; share best practice from their businesses; and build a network to make sure we’re joined up with others campaigning in this space.

We’re committed to increasing the diversity of people we represent as well. Later this year we will launch a BAME in tech network, with similar aims.

For more information and detail on the CBI’s policy thinking please read:

CBI members can also read the latest on the CBI’s Innovation and Digital policy work and get involved at MyCBI.

Published by felicityburch

Felicity Burch is the Director of Innovation and Digital at the Confederation of British Industry. She is also an adviser to the trade association, Sharing Economy UK.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: