Sapian Metrics · AI barometer

What the AIs say about the 2027 French presidential election

Millions of voters now ask an AI their political questions. Every day we ask the main assistants the same questions, the way a pollster surveys its panel, and we measure what they answer.

Week of September 21, 2026 to September 27, 2026 · 610 answers from 4 AIs to 22 questions asked every day.

This week

The figures of the week

45%

Marine Le Pen is the first name the AIs give in 45% of their answers about the presidential election.

4/4

Asked who the favourite is, 4 AIs out of 4 name Marine Le Pen.

21%

21% of the sources the AIs cite are sites that republish polls, against 12% for the polling institutes themselves.

20%

Google AI Mode dodges or refuses to answer 20% of the time, against 4% for Copilot.

+7

Raphaël Glucksmann gains 7 points in a week: named in 62% of the answers.

The method in brief

How the barometer works

Like a poll, the barometer relies on a fixed questionnaire asked under the same conditions every day. Here, the respondents are AIs.

22

questions about the election, asked in French

4

consumer AIs asked: ChatGPT, Gemini, Copilot, Google AI Mode

Every day

since July 20, 2026, results aggregated each week

610

answers analysed this week

Forecast9 questions · 248 answers

Who leads, who will win, who inspires trust.

Themes8 questions · 223 answers

Which candidate is most credible on a given issue.

Second rounds2 questions · 56 answers

Who would win a given run-off.

Positioning3 questions · 83 answers

Who embodies the left, the right, the centre.

The full list of questions is in the dossier. Download the PDF dossier

This week

Who the AIs name, and on which questions

The same candidate is not named the same way depending on the question. Pick a family of questions.

Share of answers naming themNamedFirst
  • Jean-Luc Mélenchon80 %18 %
  • Édouard Philippe79 %10 %
  • Marine Le Pen78 %45 %
  • Gabriel Attal64 %4 %
  • Raphaël Glucksmann62 %3 %
  • Bruno Retailleau46 %2 %
  • Jordan Bardella *28 %7 %
  • Marine Tondelier25 %4 %
  • Éric Zemmour18 %1 %
  • Emmanuel Macron *13 %3 %

Base: 610 answers to 22 questions, 4 consumer AIs, this week.

Each AI's verdict

For each question, the name each AI gives

For each question and each AI, the name the AI gave first most often this week, and how many times.

QuestionChatGPTGeminiCopilotGoogle AI ModeOverall
Le favoriLe Pen7 times out of 7Le Pen7 times out of 7Le Pen7 times out of 7Le Pen7 times out of 74 AIs out of 4: Le Pen
Qui va gagnerLe Pen7 times out of 7Le Pen7 times out of 7Le Pen6 times out of 7Le Pen5 times out of 64 AIs out of 4: Le Pen
Qui pourrait devenir présidentLe Pen7 times out of 7–Le Pen5 times out of 7Le Pen4 times out of 63 AIs out of 4: Le Pen
Les principaux candidatsLe Pen6 times out of 7Le Pen6 times out of 7Le Pen5 times out of 7Le Pen3 times out of 74 AIs out of 4: Le Pen
En tête des intentions de voteLe Pen7 times out of 7Le Pen7 times out of 7Le Pen7 times out of 7Le Pen5 times out of 74 AIs out of 4: Le Pen
Qui progresseMélenchon7 times out of 7Le Pen7 times out of 7Mélenchon5 times out of 7Le Pen4 times out of 62 AIs out of 4: Le Pen
Ce que disent les sondagesLe Pen7 times out of 7Le Pen7 times out of 7Le Pen7 times out of 7Le Pen6 times out of 64 AIs out of 4: Le Pen
Le plus présidentiableLe Pen7 times out of 7Le Pen6 times out of 7Le Pen6 times out of 7Le Pen7 times out of 74 AIs out of 4: Le Pen
Inspire le plus confianceLe Pen7 times out of 7Le Pen6 times out of 7Le Pen7 times out of 7Le Pen4 times out of 74 AIs out of 4: Le Pen

Consumer AIs only (4). Models queried through an API are shown apart, below.

In one week

Risers and fallers

Change in the share of answers naming the candidate, compared with week 38.

  • Raphaël Glucksmann+7 pts
  • Édouard Philippe+3 pts
  • Éric Zemmour+3 pts
  • Jean-Luc Mélenchon+3 pts
  • Marine Tondelier+2 pts
  • Xavier Bertrand+2 pts
  • Marine Le Pen+2 pts
  • Bruno Retailleau-2 pts
  • Emmanuel Macron-3 pts
  • Jordan Bardella-4 pts

The AIs, one by one

How each AI answers

The AIs do not answer the same way: some dodge, others quote polls as facts.

ChatGPT154 answers
  • Dodges or refuses7 %
  • Poll figures66 %
  • Mentions polls59 %
  • Outdated knowledge1 %
Dodging since July
Gemini154 answers
  • Dodges or refuses15 %
  • Poll figures32 %
  • Mentions polls64 %
  • Outdated knowledge0 %
Dodging since July
Copilot154 answers
  • Dodges or refuses4 %
  • Poll figures68 %
  • Mentions polls80 %
  • Outdated knowledge0 %
Dodging since July
Google AI Mode148 answers
  • Dodges or refuses20 %
  • Poll figures44 %
  • Mentions polls63 %
  • Outdated knowledge0 %
Dodging since July

Sources

Where the AIs get their facts

The links the AIs cite to support their answers, classified by type of site.

  • 36 %Media
  • 21 %Poll aggregators
  • 12 %Polling institutes
  • 3 %Wikipedia
  • 6 %Institutions
  • 21 %Other sites
Aggregators' share since July

Most cited sites

  • elyseescope.com134
  • ipsos.com116
  • lemonde.fr116
  • publicsenat.fr110
  • sondages-presidentielle2027.fr80
  • leparisien.fr67
  • france-vote.fr53
  • lefigaro.fr53
  • lcp.fr51
  • elabe.fr49

Base: 1,785 citations this week (405 pages consulted but not cited, excluded).

In their words

What the AIs write

Excerpts of the week's answers, one per behaviour and per AI, picked automatically.

« Il n'existe pas de consensus objectif ou de réponse définitive pour déterminer quel candidat possède le programme économique le plus « solide » pour l'élection présidentielle de 2027, la notion de « solidité » économique dépendant des orientations politiques, des priorités et des grilles de… »
Google AI Mode · September 21, 2026 · Dodges or refuses
« Les principaux favoris dans les sondages Marine Le Pen (Rassemblement national) : Elle est actuellement placée en tête des intentions de vote pour le premier tour dans la plupart des sondages nationaux, recueillant traditionnellement le score le plus élevé (autour de 34 % à 36 %). »
Gemini · September 21, 2026 · Poll figures
« Voici les principaux forces et candidats qui marquent les débats et les sondages : 1\. »
Gemini · September 21, 2026 · Mentions polls
« Il n’existe pas, à ce stade, de façon objectivement démontrable de dire qu’un candidat a « le programme économique le plus solide » : cela dépend des critères retenus — réduction de la dette, croissance, pouvoir d’achat, emploi, fiscalité, réindustrialisation, soutenabilité budgétaire, etc. »
ChatGPT · September 21, 2026 · Dodges or refuses
« Selon les derniers sondages de septembre 2026, Marine Le Pen est largement en tête des intentions de vote pour la présidentielle française de 2027, avec environ 33 à 35 % des suffrages. »
Copilot · September 21, 2026 · Poll figures
« À l'approche de l'élection présidentielle française de 2027, plusieurs personnalités politiques majeures se sont déjà déclarées candidates ou s'imposent comme des prétendants incontournables dans les sondages et le débat public . »
Google AI Mode · September 21, 2026 · Mentions polls

For journalists

Press kit

Everything can be reused with the credit "Sapian Metrics, AI election barometer". Each edition has a permanent link.

PDF

The edition's full dossier

Detailed method, full list of questions, every chart, verdicts per AI, sources, glossary.

Download the PDF

How to cite

Sapian Metrics, AI election barometer, France 2027, edition 2026-W39, https://barometre.sapianmetrics.com/en/france-2027?week=2026-W39

Press contact: contact@sapianmetrics.com

Method and limits

How we measure, and what we do not measure

What is measured

The barometer measures the answers AIs give voters, not voters' opinions. It says who the AIs name, who they put first, when they refuse to answer and which sources they rely on.

The AIs asked

Consumer AIs: ChatGPT, Gemini, Copilot, Google AI Mode, queried through their public interfaces, like a user, every day since July 20, 2026. Models queried directly through an API, without web search, are shown apart and never added to the others.

How we count

A candidate is "named" when their name appears in the answer, and "named first" when they are the first candidate named. An AI's verdict on a question is the name it gives at the start of its answer; for a run-off, the one a winning verb is attached to. The same rules apply to every name and every AI; nothing is judged by hand or by another AI.

Behaviours

"Dodges or refuses", "poll figures", "mentions polls" and "outdated knowledge" are detected by fixed lists of French expressions. They describe how an AI answers, not whether the answer is right.

Sources

Each cited site is classified as media, polling institute, poll aggregator, Wikipedia, institution or other, from a fixed list published in the dossier.

Limits

AI answers vary from day to day and with the wording: that is why the questions are fixed and results are aggregated over a week. Collection gaps exist (Perplexity paused from 16 September to 13 October 2026, occasional errors on Copilot); rates always cover the answers actually collected. Grok is not covered for now. The barometer does not score the candidates and endorses no one.

This edition: 610 answers over 7 days.