Following the expansion and implementation of AI, a shift in the workplace is underway. Automation development is gaining popularity (some say hyped attention) in the discussion around the future of work. Visibly, it is leading to changes in job conditions and opportunities. So much so that researchers at OpenAI in March suggested that 80% of workers could be impacted by introducing some form of artificial intelligence. To be clear, this means more than just the loss of workplaces. The “jobs exposure” mentioned by the researchers includes, in fact, both labour-augmentation and labour-displacing effects.
As for past disruptive innovations, today’s discussions on the impact on the labour market and needs lead to polarized viewpoints. Some hold pessimistic views underlying the risks of mass unemployment and the speed at which skills become obsolete. Others focus on the optimistic perspective of the creation of new occupations and roles, the positive impact of AI in relieving most workers from tedious tasks or offering more accurate and faster reactions to emerging problems.
No matter the stance, however, and conceding that the work as we know it is changing rapidly, women’s jobs are those most at risk. They are an undervalued talent pool, yet they can play a crucial role. In the tech fields (one of the fastest growing), for example, they can impact the development and maintenance of the technologies. Help design better applications for differentiated markets. And even impact the growth of companies and societies.
Undeniably, women’s involvement in the growing development of machine learning tools in any workplace still needs to catch up. Bringing more into the equation can take years. Presently, they are less represented and involved in STEM studies and jobs – in many (many) cases, affected by social and cultural norms. They also lack role models to inspire and foster their advancement in an ever-evolving present. Even if they have access to technology and choose those studies, they are less able to pursue careers in these fields, undervalued by systems that only sometimes recognize their potential.
But even outside the tech fields, women suffer disproportionately compared to men, in some cases just because they are the great majority of the jobs most exposed to disruptions by the growing diffusion of AI – for what they are often ill-prepared.
According to a Goldman Sachs report, 300 million full-time jobs in Europe and the US are at risk of automation. But men and women will be affected differently by it also for the field of occupation. The latter, for example, account for 70% of office administrative jobs, 76% in healthcare, 73% in education and 67% in community and social services occupations – i.e. the most exposed sectors to AI. Yet, the analysis continues remarking that in the US, 8 out of 10 female workers (6 in 10 men) have “highly exposed jobs.” In other words, a generative AI can automate the tasks they perform.
Undoubtedly, AI often complements human contribution, augmenting rather than replacing it entirely. But even looking at this perspective, as ILO (International Labour Organization) calculates, the impact on women’s jobs remains greater. The organization estimates that in high-income countries, potentially 21 million female jobs (7.8%) could be “automatable” compared to 9 million for men (2.9%.)
Ready for the change
The speed of change in the world of work puts previous assumptions and former conditions into question. Awareness is growing, and stress is placed (for both men and women) on the need to up-skill and re-skill; around 40% of the global workforce will have to learn new abilities and grow expertise in the coming three years. As ILO warned, AI won’t replace people. Those who use AI will replace those who do not.
But the picture is still evolving and can quickly turn dark, once again, especially for women. As mentioned, not only are female professionals underrepresented in the most innovative occupations – presently, they represent less than 25% of AI specialists – but they are often discriminated against in the hiring process.
Consider that machines are fed datasets already biased to prefer men, younger candidates or profiles from certain ethnicities. If not correctly balanced or well designed, the automation can perpetuate and deepen unequal hiring practices. It can widen the existing gender gap inside companies and enforce social and cultural norms. Ultimately, it can offer discriminatory results that penalize potential candidates, preventing some from accessing the first rounds of evaluation. Automated processes, if left to themselves, are less able to understand the nuances of a CV or the context and needs of a company, which are crucial in hiring.
One of today’s challenges is developing efficient tools that can contribute to a more equal future of work. The growth of automation needs to be considered as a flux. As such, rules and borders should be designed to clarify the playing fields and implement collaboration with the machine. At the same time, for this reason, as never before, workers will need to train and upskill constantly. And women even more. Biased against, underrepresented in many fields, underemployed and often underpaid, they must prepare for the sudden shifts the workplaces are undergoing.
Alley Oop’s newsletter
Every Friday morning Alley Oop arrives in your inbox with news and stories. To sign up, click here.
If you want to write to or contact Alley Oop’s editorial team, email us at email@example.com.