It is indeed striking how little we discuss some of these roles in the AI context vs others. The test of whether a job turns up in childrens’ books always seems like a good guide to wider public salience too…….

Looking across the categories, I wonder if it is better to consider all these roles as a group, given their silent nature + higher share of women; or better to single out specific roles, and really get a handle on how AI may affect them? Some do seem quite different to each other e.g bookeeping vs secretary (incl. more modern incarnations) vs hotel receptionists?

I think the customer service rep category is so important, given the overall trend line in your graphic, and its wider role in global employment. It also seems one of the most poorly-discussed in AI circles. (“Chatbots will automate all customer service jobs, obviously!”)

I was looking to find some evidence on customer service work and/or call centre work (which seem overlapping but also quite distinct) a couple of years ago, when people were making grand proclamations about AI automating it all. I really struggled to find anything, to your wider point here. Possibly by design!

It’s pretty clear though that the work is much more than just being able to answer questions, which it seems to be often equated to. Indeed, AI may have better prospects of automating FAQ pages than people’s roles, at least in the first instance.

The types of tasks can vary a lot, from answering a customer’s objective information-seeking question to making policy decisions to things more akin to sales/cold calling. It also seems like new technology has always historically led to more customer service demand, at least of the call center variety. And in many ways is the reason why that latter sector exists (both from an enabling point of view, and a demand point of view).

AI might change all that of course. But it also begs the question of whether AI may lead to new kinds of customer service roles (always so hard to predict).

It did make me think that we need much richer data, and stories like you have here, of what it really means to work in customer service and/or call centres, and how those role are evolving with AI.