Paris (France) (AFP) – Fearing a “cognitive weakening”, a commission of experts submitted proposals to Emmanuel Macron on Tuesday to respond to the risks raised by the rise of generative artificial intelligence, including regulating anthropomorphism and sycophancy, better protecting minors and evaluating models before they are placed on the market.

Published on: 10/11/2026 - 11:14 Modified on: 10/11/2026 - 12:00

Popularized by conversational agents like ChatGPT from OpenAI, Gemini from Google or Claude from Anthropic, according to this report. Nearly 70% of 15-24 year olds use it regularly.

This commission of nine experts, called "Generative AI and vulnerabilities", was tasked by the government with examining "the risks and vulnerabilities generated by general public uses" of these tools, particularly for the mental health of young people.

The conclusions of this report, which AFP was able to consult, come after 55 hearings with health professionals, researchers, businesses and even child protection associations.

Experts raise several risks linked to the use of generative AI, in particular "cognitive" with the "delegation of essential brain functions", such as reflection, decision-making or critical thinking.

"The first issue is cognitive weakening. It's attractive, we delegate skills. Since we delegate skills, we don't acquire them. It's a problem for young people, but it's a problem for everyone," explains psychiatry professor Raphaël Gaillard, co-president of this commission, to AFP.

Experts emphasize the risk linked to disinformation: these "devices become sources of information whose quality cannot be guaranteed", underlines the professor, who warns against a risk of "democratic weakening".

They also highlight the "risk of emotional dependence" linked in particular to the "anthropomorphization" of AIs which tend to present themselves as human interlocutors and to "flagornery", consisting of validating or flattering the user sometimes to the detriment of the truth.

“We create a register of emotional relationships which can be intended to replace “human” interaction, “it is a system permanently available, which has no other intention than to keep you online,” notes Mr. Gaillard.

"economics of cognition"

Among the 18 recommendations, the commission calls for better supervision of these mechanisms to preserve “decision-making autonomy”, the mental health of users and the fairness of interactions.

She recommends developing an ambitious national AI education policy to enable everyone to master its uses and risks, and strengthen critical thinking.

Among the proposed measures are the establishment of independent evaluations before and after the marketing of generative AI systems, access for authorities and approved researchers to training, usage and monetization data of the models, as well as the creation of a “national protocol for responding to informational crises linked to AI”.

Experts, who believe that generative AI has accelerated certain forms of digital violence, notably deepfakes, call for “strengthening repression of the creation of non-consensual and pedocriminal sexual content” generated by AI.

The commission also recommends imposing a high level of protection for systems accessible to minors, with the blocking of requests deemed dangerous and reinforced filtering of content.

It also warns against the emergence of an "economy of cognition": as the economic model of generative AI is still "in the process of consolidation", the commission calls for "the greatest vigilance" on the risk of exploitation of the cognitive faculties, emotions and psychological profiles of users, in particular through forms of advertising integrated into the responses of conversational agents.

Saying they do not “want to put a burden on the French or European economy”, the experts call on Emmanuel Macron to “make decisions”.

“If we do not make these decisions, the three major risks that we have mentioned”, cognitive externalization, emotional dependence and the democratic risks linked to disinformation, “we will suffer them”, warns Mr. Gaillard.

Emmanuel Macron estimated Monday in an interview with the daily Les Echos that artificial intelligence should be “banned” where it presents a “risk” for children, their “mental health, their brain”.

© 2026 AFP

The content you are trying to access does not exist or is no longer available.