AI unlikely to destroy jobs, but cost to certain workers may be ‘brutal’ — UN study


Generative AI is extra more likely to complement current jobs than take over them fully, although sure roles corresponding to clerical work — might see extra of their duties automated than others. 

According to an Aug. 21 Generative AI and Jobs research by the Worldwide Labour Group (ILO) — a United Nations company — 24% of clerical duties are thought-about extremely uncovered to automation, with a further 58% with medium-level publicity.

Duties with excessive and medium GPT-exposure by occupational class. Supply: ILO

Typists, journey consultants, financial institution tellers, contact heart clerks, bookkeeping and information entry clerks, lodge receptionists and secretaries are the administration roles most in danger, the figures present.

This, in accordance with the ILO, might recommend that ladies could possibly be extra in danger, given their larger representation in administrative roles.

“3.7 per cent of all feminine employment on the earth is in jobs which can be doubtlessly automatable with generative AI know-how, in contrast with only one.4 per cent of male employment.”

In the meantime, AI automated work is extra more likely to impression staff in high-income nations (5.5%) in comparison with low-income nations (0.4%), the report discovered:

Occupations with excessive automation potential by revenue stage and intercourse. Supply: ILO

The ILO’s research on generative AI largely targeted on the impression of chatbot functions, corresponding to OpenAI’s ChatGPT and Google’s Bard.

Crypto customer support

The ILO report additionally reveals customer support and coordination-related duties as having excessive automation potential, together with information administration and report holding, data processing and language providers, duties associated to responding to inquiries. 

Desk exhibiting duties with excessive automation potential clustered into thematic teams. Supply: ILO

Many customer support roles have been misplaced in the newest crypto winter of 2022, which noticed a number of the trade’s heavyweights in Binance, Coinbase and Kraken considerably cut back headcounts, together with customer support.

At the moment, customer support roles in Web3 comprise 832 (2.5%) of the overall 33,846 listings on cryptocurrency job board Web3.career.

Associated: Dear crypto writers: No one wants to read your ChatGPT-generated trash

Nonetheless, the ILO concluded that the workforce as an entire will not be too affected by AI and that AI’s total impression have been neither significantly constructive nor damaging for now — reasonably, its impression will depend upon how GPTs are managed and controlled.

“With out correct insurance policies in place, there’s a threat that solely a number of the well-positioned nations and market members will be capable of harness the advantages of the transition, whereas the prices to affected staff could possibly be brutal,” it wrote. 

ILO’s findings are extra optimistic than that of on a regular basis People, with a latest survey revealing that 62% of the U.S. inhabitants consider AI will have a major impact in the workplace over the subsequent 20 years, leaving many People “cautious” and “anxious” about what their future holds.

Journal: AI Eye: AI’s trained on AI content go MAD, is Threads a loss leader for AI data?