Conspiracy theorist Alex Jones seized on a May 9 video of a train car meeting among three European leaders to claim they had used drugs and were trying to hide it. The video showed French President Emmanuel Macron sitting at a table with German Chancellor Friedrich Merz and United Kingdom Prime Minister Keir Starmer. On the table before them were two blue folders, two drinking glasses and a small white object. The three men smiled for photographers who had gathered. Just as the shutter clicks started, Macron removed the crumpled white object from the tabletop and held it in his fist.

We present both the Lucie Training Dataset and the Lucie-7B foundation model. The Lucie Training Dataset is a multilingual collection of textual corpora centered around French and designed to offset anglo-centric biases found in many datasets for large language model pretraining. Its French data is pulled not only from traditional web sources, but also from French cultural heritage documents, filling an important gap in modern datasets. Beyond French, which makes up the largest share of the data, we added documents to support several other European languages, including English, Spanish, German, and Italian. Apart from its value as a resource for French language and culture, an important feature of this dataset is that it prioritizes data rights by minimizing copyrighted material. In addition, building on the philosophy of past open projects, it is redistributed in the form used for training and its processing is described on Hugging Face and GitHub. The Lucie-7B foundation model is trained on equal amounts of data in French and English -- roughly 33% each -- in an effort to better represent cultural aspects of French-speaking communities. We also describe two instruction fine-tuned models, Lucie-7B-Instruct-v1.1 and Lucie-7B-Instruct-human-data, which we release as demonstrations of Lucie-7B in use. These models achieve promising results compared to state-of-the-art models, demonstrating that an open approach prioritizing data rights can still deliver strong performance. We see these models as an initial step toward developing more performant, aligned models in the near future. Model weights for Lucie-7B and the Lucie instruct models, along with intermediate checkpoints for the former, are published on Hugging Face, while model training and data preparation code is available on GitHub. This makes Lucie-7B one of the first OSI compliant language models according to the new OSI definition.

This study examines ChatGPT-4's capability to replicate linguistic strategies used in political discourse, focusing on its potential for manipulative language generation. As large language models become increasingly popular for text generation, concerns have grown regarding their role in spreading fake news and propaganda. This research compares real political speeches with those generated by ChatGPT, emphasizing presuppositions (a rhetorical device that subtly influences audiences by packaging some content as already known at the moment of utterance, thus swaying opinions without explicit argumentation). Using a corpus-based pragmatic analysis, this study assesses how well ChatGPT can mimic these persuasive strategies. The findings reveal that although ChatGPT-generated texts contain many manipulative presuppositions, key differences emerge in their frequency, form, and function compared with those of politicians. For instance, ChatGPT often relies on change-of-state verbs used in fixed phrases, whereas politicians use presupposition triggers in more varied and creative ways. Such differences, however, are challenging to detect with the naked eye, underscoring the potential risks posed by large language models in political and public discourse.Using a corpus-based pragmatic analysis, this study assesses how well ChatGPT can mimic these persuasive strategies. The findings reveal that although ChatGPT-generated texts contain many manipulative presuppositions, key differences emerge in their frequency, form, and function compared with those of politicians. For instance, ChatGPT often relies on change-of-state verbs used in fixed phrases, whereas politicians use presupposition triggers in more varied and creative ways. Such differences, however, are challenging to detect with the naked eye, underscoring the potential risks posed by large language models in political and public discourse.

Vice President JD Vance addressed the AI Action Summit in Paris Tuesday during his first foreign trip since taking office. OpenAI CEO Sam Altman commended Vice President JD Vance's artificial intelligence (AI) speech in Paris on Tuesday while laying out his vision for how people can take advantage of the rapidly evolving technology at the same conference. Altman and Vance appeared Tuesday at the AI Action Summit in Paris, where world leaders, top tech executives and policymakers teamed up to hash out tech policy and its intersection with global security, economics and governance. During his remarks, Vance called for AI systems developed in the U.S. to remain free of "ideological bias" and vowed that the U.S. would "never restrict our citizens' right to free speech." Vance also pushed for a "deregulatory flavor" to emerge at the conference while cautioning against the pitfalls of "excessive regulation" that could hamper a transformative industry.

U.S. Vice President JD Vance said that the Trump administration will work to make the U.S. the "gold standard worldwide" for artificial intelligence as he issued strong warnings against regulating political speech. Speaking Tuesday to an audience in Paris that included several European Union leaders, he took particular aim at the bloc's tough regulatory approach to social media platforms and AI, accusing it of trying to clamp down on Silicon Valley. "The Trump administration is troubled by reports that some foreign governments are considering tightening the screws on U.S. tech companies with international footprints," Vance said during an AI summit hosted by French President Emmanuel Macron. "Now America cannot and will not accept that, and we think it's a terrible mistake, not just for the United States of America, but for your own countries."

Europe will cut back on regulation to make it easier for artificial intelligence to flourish in the region, French President Emmanuel Macron told an AI summit in Paris on Monday, urging investment in the EU -- and more specifically in France. The European Union's digital chief Henna Virkkunen also promised that the bloc will simplify its rules and implement them in a business-friendly way. As U.S. President Donald Trump has torn up his predecessor's AI guardrails to boost U.S. competitiveness, pressure has built on the EU to pursue a lighter-touch approach to AI regulation to help keep European companies in the technology race.

Decoding strategies for generative large language models (LLMs) are a critical but often underexplored aspect of text generation tasks. Guided by specific hyperparameters, these strategies aim to transform the raw probability distributions produced by language models into coherent, fluent text. In this study, we undertake a large-scale empirical assessment of a range of decoding methods, open-source LLMs, textual domains, and evaluation protocols to determine how hyperparameter choices shape the outputs. Our experiments include both factual (e.g., news) and creative (e.g., fiction) domains, and incorporate a broad suite of automatic evaluation metrics alongside human judgments. Through extensive sensitivity analyses, we distill practical recommendations for selecting and tuning hyperparameters, noting that optimal configurations vary across models and tasks. By synthesizing these insights, this study provides actionable guidance for refining decoding strategies, enabling researchers and practitioners to achieve higher-quality, more reliable, and context-appropriate text generation outcomes.

Generative AI proposes several large language models (LLMs) to automatically generate a message in response to users' requests. Such scientific breakthroughs promote new writing assistants but with some fears. The main focus of this study is to analyze the written style of one LLM called ChatGPT by comparing its generated messages with those of the recent French presidents. To achieve this, we compare end-of-the-year addresses written by Chirac, Sarkozy, Hollande, and Macron with those automatically produced by ChatGPT. We found that ChatGPT tends to overuse nouns, possessive determiners, and numbers. On the other hand, the generated speeches employ less verbs, pronouns, and adverbs and include, in mean, too standardized sentences. Considering some words, one can observe that ChatGPT tends to overuse "to must" (devoir), "to continue" or the lemma "we" (nous). Moreover, GPT underuses the auxiliary verb "to be" (^etre), or the modal verbs "to will" (vouloir) or "to have to" (falloir). In addition, when a short text is provided as example to ChatGPT, the machine can generate a short message with a style closed to the original wording. Finally, we reveal that ChatGPT style exposes distinct features compared to real presidential speeches.

This paper aims to provide a comparison between texts produced by French and Italian politicians on polarizing issues, such as immigration and the European Union, and their chatbot counterparts created with ChatGPT 3.5. In this study, we focus on implicit communication, in particular on presuppositions and their functions in discourse, which have been considered in the literature as a potential linguistic feature of manipulation. This study also aims to contribute to the emerging literature on the pragmatic competences of Large Language Models. Our results show that, on average, ChatGPT-generated texts contain more questionable presuppositions than the politicians' texts. Furthermore, most presuppositions in the former texts show a different distribution and different discourse functions compared to the latter. This may be due to several factors inherent in the ChatGPT architecture, such as a tendency to be verbose and repetitive in longer texts, as exemplified by the occurrence of political slogans mainly formed by change-of-state verbs as presupposition triggers (e.g., dobbiamo costruire il nostro futuro, 'we must build our future').

From fake images designed to cause fears of an immigrant "invasion" to other demonisation campaigns targeted at leaders such as Emmanuel Macron, far-right parties and activists across western Europe are at the forefront of the political weaponisation of generative artificial intelligence technology. This year's European parliamentary elections were the launchpad for a rollout of AI-generated campaigning by the European far right, experts say, which has continued to proliferate since. This month, the issue reached the independent oversight board of Mark Zuckerberg's Meta when the body opened an investigation into anti-immigration content on Facebook. The inquiry by the oversight board will look at a post from a German account featuring an AI-generated image emblazoned with anti-immigrant rhetoric. It is part of a wave of AI-made rightwing content on social media networks.