Data study ·
AI at work: employees are ready, companies are not
Original title, in French: “IA au travail : les salariés sont prêts, pas les entreprises”
- Publication
- Comarketing-News (France)
- Author
- Unsigned
- Published on
- 27 October 2026 · read in the page structured data
Summary
A study by Cognizant Research, conducted in early 2026 among more than 10,000 employees in 24 countries including France, reveals an AI "activation gap" in the workplace: 80% of employees want to use AI more, but tools and organisation are not keeping up. Between 72% and 82% of employees consider the tools poorly suited to their needs, and 87% of the most advanced users ("Architects") use unauthorised tools. The study identifies seven employee profiles and shows that individual dispositions predict adoption far better than position in the hierarchy. According to Cognizant, a 5% increase in global productivity would generate around $6 trillion of additional GDP, but this potential remains held back by tools that lack contextualisation and training that lacks personalisation.
Summary based on the full text of the article.
Read the article on Comarketing-News →Figures cited
-
80 %
Share of employees who want to use AI more than their role allows
more than 10,000 employees in 24 countries, including France, early 2026
« quatre salariés sur cinq souhaitent utiliser davantage l’IA que ne le leur permet actuellement leur poste »
Source: Cognizant Research, étude New Frontiers, New Frictions: Mapping the DNA of AI Adoption
-
6,000 USD billion
Additional global GDP generated by a 5% increase in labour productivity
global estimate, Cognizant Research
« une augmentation généralisée de 5 % de la productivité du travail à l’échelle mondiale pourrait générer environ 6 000 milliards de dollars de PIB supplémentaire »
Source: Cognizant Research
-
82 %
Share of employees who consider the AI tools provided to them do not meet their needs
employees surveyed, by profile (range 72-82%)
« Entre 72 % et 82 % des salariés, selon leur profil, estiment que les outils d’IA mis à leur disposition ne répondent pas à leurs besoins »
Source: Cognizant Research
-
28 %
Share of Perfectionists satisfied with the available AI solutions
Perfectionist profile, Cognizant study
« parmi les Perfectionnistes, le groupe le plus satisfait, seuls 28 % se déclarent satisfaits des solutions disponibles »
Source: Cognizant Research
-
33 %
Share of Accelerators and Architects citing lack of contextualisation as the main barrier
Accelerator and Architect profiles, Cognizant study
« Parmi les Accélérateurs, utilisateurs pragmatiques et réguliers, et les Architectes, pionniers de l’adoption, 33 % identifient le manque de contextualisation et de personnalisation comme le principal frein à une utilisation plus poussée »
Source: Cognizant Research
-
87 %
Share of Architects using AI tools not authorised by their company
Architect profile, Cognizant study
« 87 % des Architectes déclarent recourir à des outils d’IA non autorisés par leur entreprise lorsque les solutions mises à leur disposition ne répondent pas à leurs attentes »
Source: Cognizant Research
-
72 %
Share of the workforce studied represented by the Navigator and Pillar profiles
Cognizant study sample
« les Navigateurs et les Piliers représentent à eux seuls 72 % des effectifs étudiés »
Source: Cognizant Research
-
81 %
Share of Architects who consider their AI training effective
Architect profile, Cognizant study
« Parmi les Architectes, 81 % jugent leur formation à l’IA efficace et 83 % la suivent jusqu’à son terme »
Source: Cognizant Research
-
56 %
Share of Resisters who consider their AI training effective
Resister profile, Cognizant study
« Chez les Réfractaires, ces proportions chutent respectivement à 56 % et 53 % »
Source: Cognizant Research
Classification
- Country
- France
- Company size
- All sizes
- Industries
- All industries, or industry not specified
- Companies
- Cognizant (study author)
- Keywords
- AI adoption, employee training, organisational barrier, unauthorised AI tools, productivity, Cognizant study
Entry written by Marsanne Studio from the cited article, with the help of an AI model (claude-sonnet-5). The rights to the article and its illustrations belong to Comarketing-News. Entry translated from French by Claude Sonnet 5.5.