Employment and skills: what automation changes

Read the PDF edition (2 pages)

Artificial intelligence affects the substance of the financial centre’s professions, which raises the question of employment. The most exposed functions are those whose content is repetitive and documentary: transaction processing, data entry and verification, report production, the first-level analysis of compliance alerts, the drafting of standardised documents and routine translation. The least exposed functions are those in which relationships, judgement and responsibility predominate: advising complex clients, negotiation, credit decisions, representation before the authorities and management. Between the two, most professions will see their content shift towards the supervision of tools, the interpretation of their results and the handling of the cases that the tools do not resolve.

For a financial centre of 38,000 jobs, a significant share of them in support and control functions, this development could reduce headcount in some categories and create it in others. The experience of previous technological waves suggests that the financial centre’s total employment depends more on its volume of business than on its productivity.¹ Skills, on the other hand, will change: institutions will need staff able to use the tools, to understand their limits, to verify their results and to retain, in their own field, the mastery that the tools do not replace. This point, on which Part J insisted with regard to language, applies equally to law, taxation, financial analysis and compliance.

Worked example. The compliance department of a Geneva bank handles 12,000 transaction monitoring alerts a year, 95 per cent of which prove to be unfounded after a manual analysis of roughly thirty minutes each, that is, 5,700 hours of work. An analysis tool based on machine learning, validated in accordance with FINMA’s expectations and supervised by the analysts, automatically dismisses 60 per cent of the unfounded alerts with a documented error rate lower than that of manual analysis. The department saves around 3,400 hours, which it reallocates to the in-depth analysis of complex cases and the review of high-risk relationships. Headcount is maintained, but the content of the work changes. The bank invests in training its analysts to supervise the tool.

Notes

1. “La place financière genevoise "se porte bien"” (The Geneva financial centre “is doing well”), Léman Bleu, 16 October 2025: nearly 38,000 jobs in Geneva’s financial sector. ↩

For general information only. This article does not constitute legal or tax advice, does not replace advice on an individual case and creates no lawyer-client relationship. It reflects the legal position as of September 2025.