Why Women Must Engage with Generative AI (and How to Get Started)

Study after study after study shows that fewer women than men are using generative AI. 

The reasons for this AI gender gap are numerous and, to be frank, are not new. Some of the causes mentioned in the emerging literature are:

• Lack of exposure to generative AI in the workplace: There is still a difference in terms of the type of jobs that men do vs women. The lack of exposure results in fewer opportunities to use generative AI.
• The confidence gap: Women often underestimate their technical skills, even when they are just as competent as men. They also feel that they need training in order to use AI. As a result, they are not putting themselves forward for tasks or projects that might involve the use of this technology.
• Privacy and ethical concerns: Some studies suggest that women are more concerned than men about data privacy and security. Others, that women are more concerned than men with the ethical issues surrounding the use of generative AI. These concerns could make women more hesitant to engage with AI tools, where there is not a clear benefit from doing so.
• Opportunity cost: Women are still having to do more of the household and family chores. They are also more likely than men to be penalised for making mistakes in public and in the workplace. Thus, experimenting with generative AI could be more “costly” for women than men.

Why the gender gap matters (and to whom)

The lack of women’s engagement with generative AI is, first and foremost, a problem for women themselves. Generative AI related skills are increasingly a requirement in job adverts. In some cases, Generative AI skills are even more important than job experience. So, failing to use Gen AI and develop AI skills is having an impact on women’s employability. Moreover, the technology is becoming more and more sophisticated, and many AI-related jobs (e.g., AI ethics, AI strategy) are emerging. However, if women don’t engage now and do not develop AI literacy and skills, they risk missing out on these career paths and hurting their earning potential. It will also become even harder for female entrepreneurs to secure funding. It is already harder for women to secure VC funding than for men; and early evidence shows that gap is even wider for women funded AI start-ups.

But the gender gap in generative AI is a concern for organisations and society, too. By not engaging with this technology and not joining in the associated strategic discussions, we are not part of the decisions that shape the nature and use of Gen AI technology. Who remembers the time when Apple launched the HealthKit to give individuals a single comprehensive picture of their health but then forgot to include a period tracker? There is also the issue of representation and visibility: the fewer women are using and discussing generative AI, the more it may feel like a boys-club, further deterring women from engaging with it. And, of course, generative AI users can help fine tune the underpinning models through its queries or by providing feedback on its outputs by thumbs up or down. They can flag, for instance, that a translation from English to Portuguese presented doctors as male but nurses as female, or that generative AI-generated stories about men include terms such as treasure or adventure while stories about women favour terms such as gentle, love or husband, for no other reason than embedded bias. 

What you can do

As a manager or decision maker, you can’t rely on people coming to you with those skills, already. If you do, you will face a pool of applicants, from the start, that is less concerned with privacy and security risks, and you will end up lacking a diversity of views and experiences (see Apple example, above). Invest in training – it is good for employers. And, when you do, make sure that you do not limit it to technical roles, and that you offer a variety of training options (e.g., structured training with clear use cases, as opposed to trial and error). This ensures that women aren’t left behind.

But, as an individual, you need to invest in your own training, too. Here are some suggestions, not just for women:

• Experiment, experiment, experiment – This is like riding a bike: you can’t learn just from reading about it. Yes, there are many useful guides online, like Coursera’s Google AI essentials. But, at the end of the day, you need to do it to learn it. Start small, reflect on your experiences, try again.
• Make it relevant – Yes, generative AI is a complex technology. But that doesn’t mean that you need to use it for complex matters, only. Use it for what interests you, and you would find valuable. For instance, planning an event, writing a difficult letter or coming up with examples for your teaching.
• Use it for things that you are good at – Generative AI sounds convincing, but it is not a knowledge model. Sometimes (many times), the output is just botshit. By using generative AI it in topics and activities that you are very good at, it will be easier for you to understand when it makes sense to use it or not.
• Flag biases and mistakes – When you come across biases and mistakes, challenge them through the dialogue boxes, or simply use the feedback buttons (thumbs up/down) to correct them. Small actions will add up to make AI better for everyone.
• Share experiences – Ask people around you how they are using generative AI, and what they are finding it useful for. If they are ahead of you in terms of skills and experience, you can learn from them. And if they are behind you, you can help close the AI gender gap.

What’s one small step you can take today to close the AI gender gap?