The ways AI will change the future of work

Since the arrival of Chat­G­PT, there has been con­cern that we are going to be replaced by robots. Gen­er­at­ive arti­fi­cial intel­li­gence, cap­able of assim­il­at­ing and cre­at­ing writ­ten, visu­al, or audio con­tent, is often described as a threat to jobs. With each new major tech­no­lo­gic­al advance comes its share of debate and appre­hen­sion about impacts on the work­force. Dur­ing the indus­tri­al revolu­tion, manu­al work­ers were at the fore­front of these major changes. Con­versely, AI is now more rel­ev­ant to man­agers and pro­fes­sion­als. But what will be the real impact of this technology?

Togeth­er with Pawel Gmyrek and Dav­id Bescond, Inter­na­tion­al Labour Organ­isa­tion eco­nom­ist Jan­ine Berg ana­lysed the 436 occu­pa­tions lis­ted in the ILO’s Inter­na­tion­al Clas­si­fic­a­tion of Occu­pa­tions. The aim was to under­stand which types of jobs would be most affected by AI on a glob­al scale. The authors used Chat­G­PT to ana­lyse the tasks asso­ci­ated with the occu­pa­tions and assigned them scores cor­res­pond­ing to their poten­tial expos­ure. Some tasks are highly exposed to tech­no­logy, oth­ers less so. The high­er the expos­ure poten­tial of an activ­ity, the more likely it is to be automated.

For eco­nom­ists, the primary impact of arti­fi­cial intel­li­gence will not really be the massive destruc­tion of jobs, but rather the pro­found trans­form­a­tion of work. For most pro­fes­sions, cer­tain tasks will indeed be car­ried out by bots, but this will leave time for oth­er, more com­plex activ­it­ies. On aver­age, 10–13% of jobs world­wide could be “aug­men­ted” or trans­formed. The first jobs to use this tech­no­logy will poten­tially be ware­house­men, deliv­ery drivers, man­agers in the dis­tri­bu­tion sec­tor, machine oper­at­ors and assem­blers, ser­vice and sales work­ers, driv­ing instruct­ors, drivers, waiters, archi­tects, teach­ers, musi­cians, etc. In all, 427 mil­lion jobs, or 13% of jobs world­wide, could change because of arti­fi­cial intelligence.

Although the poten­tial for change is much great­er than auto­ma­tion, the risk remains very real, with 2.3% of jobs world­wide affected. Admin­is­trat­ive jobs would be heav­ily impacted by auto­ma­tion. “Call centre employ­ees, sec­ret­ar­ies, data entry oper­at­ors – simple, lin­ear activ­it­ies with little vari­ation in tasks and little inter­ac­tion with oth­ers – could be replaced by bots,” explains Jan­ine Berg. In recent years, office work­ers have already seen their day-to-day work evolve. Accord­ing to the experts, 24% of their tasks are highly exposed to AI, and 58% are mod­er­ately exposed. This is by far the occu­pa­tion most at risk. This means that 2.3% of jobs world­wide, or 75 mil­lion, could end up being automated.

Our study should not be read as a reas­sur­ing voice, but rather as a call to devel­op meas­ures to deal with the immin­ent tech­no­lo­gic­al changes

Arti­fi­cial intel­li­gence will clearly not affect all pro­fes­sions in the same way. The tech­no­logy is also likely to have dif­fer­ent con­sequences for men and women. Women will be 2.5 times more affected by auto­ma­tion than men, not least because there are more women in low-skilled admin­is­trat­ive pos­i­tions. Con­versely, male-dom­in­ated pro­fes­sions such as secur­ity, trans­port and con­struc­tion are unlikely to be affected. As a res­ult, 3.7% of women’s jobs world­wide are at risk of being auto­mated, com­pared with 1.4% of men’s jobs. This dif­fer­ence is even great­er in rich coun­tries, with 7.8% of jobs held by women likely to be replaced by bots, com­pared with 2.9% of jobs held by men. In low-income coun­tries, few­er women are in the labour mar­ket, and low-skilled admin­is­trat­ive pro­fes­sions are pre­dom­in­antly held by men.

The oth­er major dif­fer­ence poin­ted out by the ILO eco­nom­ists depends on the wealth of the coun­tries con­cerned. “In low-income coun­tries, arti­fi­cial intel­li­gence is unlikely to be deployed. The tech­no­logy is expens­ive, and there is a lack of infra­struc­ture, with a poor elec­tri­city sup­ply and poor inter­net con­nec­tion”, explains Jan­ine Berg. In fact, by 2022, a third of the world’s pop­u­la­tion will not have access to the inter­net. What’s more, the struc­ture of the labour mar­ket in low-income coun­tries makes them less sus­cept­ible to auto­ma­tion. In these coun­tries, 0.4% of jobs could be replaced by bots, com­pared with 5.5% in high-income coun­tries. In terms of pos­sible changes to jobs, 10.4% of occu­pa­tions are affected in low-income coun­tries, com­pared with 13.4% in rich coun­tries. In short, poten­tial auto­ma­tion mainly con­cerns rich coun­tries. They will be more dis­rup­ted by AI, but they will also be able to take advant­age of it. “This situ­ation could cre­ate a pro­ductiv­ity divide between rich and poor coun­tries”, says the economist.

While there are nuances depend­ing on the region of the world, the study broadly envis­ages the integ­ra­tion of AI into every­day life. Repla­cing humans with bots is not on the cards for the time being. “This approach could have been expec­ted to gen­er­ate an alarm­ing num­ber of job losses, but is not the case. Our over­all estim­ate points more towards a future where work is in fact trans­formed, but still present”, the eco­nom­ists sum­mar­ise. How­ever, this evol­u­tion of work must take cer­tain issues into account to avoid a neg­at­ive impact. “Our study should not be read as a reas­sur­ing voice, but rather as a call to devel­op meas­ures to deal with the immin­ent tech­no­lo­gic­al changes”, explain the authors.

Jan­ine Berg believes that gen­er­at­ive AI is fun­da­ment­ally neither pos­it­ive nor neg­at­ive. It all depends on how the tech­no­logy is deployed. The eco­nom­ist details a num­ber of actions that gov­ern­ments need to take: “reflect­ing on the ques­tion of the bal­ance of power, the voice of work­ers affected by labour mar­ket adjust­ments, respect for exist­ing stand­ards and rights, and the appro­pri­ate use of nation­al social pro­tec­tions, as well as train­ing sys­tems will be cru­cial ele­ments in steer­ing the deploy­ment of AI in the world of work.”

It is not just a ques­tion of observing how this new tech­no­logy is applied, but of sup­port­ing it with reflec­tion and meas­ures. The aim is to ensure social dia­logue, redeploy­ment or train­ing for employ­ees affected by auto­ma­tion, and employ­ee par­ti­cip­a­tion in the intro­duc­tion of AI for those whose tasks will be trans­formed. “If we don’t put meas­ures in place, and these sys­tems arrive, more jobs than neces­sary will be lost. Work­ing con­di­tions will deteri­or­ate. There may be short-term gains for some com­pan­ies, but there will be social con­sequences,” warns Jan­ine Berg.

Sirine Azouaouis