For experienced executives over 45, artificial intelligence is more than a question of technology. It calls for a reassessment of their contribution, recognition and prospects: what remains distinctive, what needs updating, what can be automated and which new responsibilities may emerge. The challenge is to renew the value of their experience as work evolves.
The launch of Impulsion IA 2027 on 22 September 2026 makes this issue particularly timely. The French government programme combines training, business support and social dialogue, with 12 announced measures, over €80 million earmarked for the coming year and an ambition to raise AI awareness or provide training to more than 30 million people. These are policy commitments whose effects will need to be assessed over time.
When AI can produce a summary, analysis or recommendation in seconds, what still makes twenty or thirty years of experience truly valuable?
Professional experience is not a single asset: technical expertise, sector knowledge, organisational understanding, customer relationships, exception handling and judgement do not all evolve in the same way.
15%more issues resolved per hourCustomer support • QJE 2025
+19%more time = SLOWERMETR • expert developers 2025
Three separate studies involving different contexts, tasks and populations. Their results are not directly comparable.
What is losing its scarcity
Where a professional’s value once rested partly on finding information quickly, applying a familiar framework or producing an initial analysis, AI can diminish that advantage. Some knowledge becomes more accessible; other knowledge needs updating; still other forms grow more valuable because they help identify exceptions, provide context and make sound judgements.
Productivity research calls for caution. In customer support, AI assistance increased the average number of issues resolved per hour by 15%, with larger gains among less experienced workers. In three trials involving developers, access to a coding assistant was associated with approximately 26% more tasks completed, again with greater benefits for less experienced participants.
Conversely, in a field study of highly experienced developers, using AI tools increased the time needed to complete the observed tasks by 19%. AI does not automatically improve an expert’s performance.
Experience is renewed through judgement
Experience can help spot inconsistencies, put an answer into context or identify an exception. But experience alone is no guarantee of better judgement.
Professional judgement means assessing the situation, spotting missing information, comparing options, recognising the limits of one’s knowledge and knowing when to decide or escalate.
AI makes this skill more demanding: a polished answer can be wrong; a coherent recommendation can rely on poor data; a procedure that was once appropriate may no longer fit the current context.
Who will develop tomorrow’s experts?
Much professional learning has historically been built on relatively simple tasks: initial research, reconciliation, review, first drafts and routine customer requests. These tasks are productive, but they are also formative: they teach people to identify anomalies, structure their reasoning and understand the consequences of mistakes.
If AI directly produces the output of a task, it does not automatically provide the learning that task once delivered. A company can save time today while weakening its ability to develop tomorrow’s experts.
The answer is not to preserve every old task artificially. Learning must be redesigned: an initial independent analysis, comparison with AI output, unfamiliar cases, expert feedback, progressively higher-stakes decisions and a review of real-world outcomes.
Experienced executives can play an active role in this new knowledge-transfer process while continuing to learn themselves.
ILLUSTRATIVE CASE — COMPOSITE SCENARIO
A head of management control, aged 52
Part of her monthly work involves consolidating data, preparing initial commentary and producing presentation materials. An AI assistant now handles preliminary data aggregation and suggests summaries.
Her role evolves: she defines alert thresholds, challenges anomalies, compares data with operational developments, decides what deserves the executive committee’s attention and helps transfer knowledge to junior controllers.
The intended benefit for the company is not merely time saved, but better allocation of expertise. Freed-up time creates value when it is reinvested in analysis, decisions and learning.
Expected result: a redefined role, explicit responsibilities and an organised transfer of knowledge rather than simply automating tasks.
Training for real work
A company can hardly conclude that an employee cannot adapt to AI if it has not provided the right tools, relevant use cases or time to learn.
Serious professional training goes far beyond learning to use a chatbot. It includes understanding limitations, mastering approved tools, applying them to specific jobs, redesigning processes, governance, data security and verification skills.
Schneider Electric, for example, describes applied workshops, “promptathons” and a network of AI advocates around its Data & AI School. The issue is not just how many people have been trained, but what they can actually do differently at work afterwards.
Age and experience: two distinct realities
Age should not be confused with experience. A younger person may be highly experienced in one activity, while an older professional may be new to a tool or occupation.
There is no robust evidence that AI experiments are systematically reserved for younger employees. Companies should therefore examine the facts: who has access to tools, pilot projects, training time and new assignments?
AXA has formalised dialogue around AI and employee-representative training; Schneider Electric has piloted a Group-level committee dedicated to AI and social dialogue. AI is therefore as much a career-development issue as a technological one.
Five questions for HR directors
Which tasks are actually changing in the work of the executives concerned?
Which skills are becoming less distinctive, and which are becoming more critical?
Do employees have the time, tools and job-relevant cases they need to learn?
What new responsibility or assignment puts these upgraded skills to practical use?
How should quality, learning and value created be measured—not just user numbers?
CONCLUSION
AI is transforming work.Human experience must become a strategic advantage.
The real challenge is not simply to train experienced executives to use new tools, but to renew their contribution, responsibilities and future prospects.
1EXPERIENCE
Human experience retains strategic value in the age of AI, provided it continues to evolve.
2COMPANY
Companies must support this renewal rather than leave experienced executives to face transformation alone.
3WENOLD
WeNold provides a structured response through career repositioning, knowledge transfer and new professional opportunities.
What WeNold support should deliver in practice
Observable outcomes, not just a general ambition.
AREA
DESIRED OUTCOME
REPOSITIONING
New assignments or responsibilities identified
SKILLS
Job-specific AI development plan
KNOWLEDGE TRANSFER
Critical knowledge identified and passed on
PERFORMANCE
Contribution assessed against tangible objectives
ENGAGEMENT
Clearer professional prospects
WeNold identifies and unlocks the Human Capital of respected, committed and experienced executives aged 45+ to turn their expertise into a new source of value for both their companies and their futures.
FROM INSIGHT TO ACTION: THE WENOLD APPROACH
WeNold helps companies and experienced executives over 45 identify, unlock and redeploy their Human Capital.
Career repositioning, knowledge transfer or the development of a new life project: our approach turns accumulated experience into new opportunities for both the company and the individual.
Selected sources directly used in this analysis. Findings are presented within their proper scope: a company case demonstrates the existence of a programme, not its causal effectiveness.
French Ministry of Labour — Impulsion IA 2027, September 2026Online source
ILO — Generative AI and Jobs: A Refined Global Index of Occupational Exposure, 20 May 2025Online source
Brynjolfsson, Li & Raymond — Generative AI at Work, QJE, 4 February 2025Online source
Cui et al. — The Effects of Generative AI on High-Skilled Work, Management Science, 27 February 2026Online source
METR — Impact of AI on experienced open-source developers, 10 July 2025Online source
Goh et al. — Large Language Model Influence on Diagnostic Reasoning, JAMA Network Open, 28 October 2024Online source
Vaccaro, Almaatouq & Malone — When combinations of humans and AI are useful, Nature Human Behaviour, 28 October 2024Online source
AI Act Service Desk — Article 14, human oversightOnline source
Schneider Electric — Data & AI School / AI upskilling, 17 July 2025Online source
Schneider Electric — social dialogue and AI deployment, 1 September 2026Online source
AXA / signatory organisations — RSG agreement on social dialogue and AI, 13 June 2025Online source
Methodological note: the productivity figures cited are context-specific results, tied to particular tasks, tools and populations. They do not represent a universal average of AI’s effects.
