{"id":2,"date":"2026-03-11T09:33:22","date_gmt":"2026-03-11T09:33:22","guid":{"rendered":"http:\/\/chatgptmatuer.fr\/?page_id=2"},"modified":"2026-03-14T08:14:09","modified_gmt":"2026-03-14T08:14:09","slug":"sample-page","status":"publish","type":"page","link":"https:\/\/chatgptmatuer.fr\/index.php\/sample-page\/","title":{"rendered":"Introduction de l&rsquo;ouvrage"},"content":{"rendered":"\n<figure class=\"wp-block-image size-large\"><img loading=\"lazy\" decoding=\"async\" width=\"683\" height=\"1024\" src=\"http:\/\/chatgptmatuer.fr\/wp-content\/uploads\/2026\/03\/ChatGPT-Image-Mar-10-2026-11_23_44-AM-2-683x1024.png\" alt=\"\" class=\"wp-image-178\" srcset=\"https:\/\/chatgptmatuer.fr\/wp-content\/uploads\/2026\/03\/ChatGPT-Image-Mar-10-2026-11_23_44-AM-2-683x1024.png 683w, https:\/\/chatgptmatuer.fr\/wp-content\/uploads\/2026\/03\/ChatGPT-Image-Mar-10-2026-11_23_44-AM-2-200x300.png 200w, https:\/\/chatgptmatuer.fr\/wp-content\/uploads\/2026\/03\/ChatGPT-Image-Mar-10-2026-11_23_44-AM-2-768x1152.png 768w, https:\/\/chatgptmatuer.fr\/wp-content\/uploads\/2026\/03\/ChatGPT-Image-Mar-10-2026-11_23_44-AM-2.png 1024w\" sizes=\"auto, (max-width: 683px) 100vw, 683px\" \/><\/figure>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>FR<\/strong><\/p>\n\n\n\n<h1 class=\"wp-block-heading\"><a><\/a><a>ChatGPT m\u2019a tuer&nbsp;!<\/a><\/h1>\n\n\n\n<h1 class=\"wp-block-heading\">Propos liminaire<\/h1>\n\n\n\n<p class=\"wp-block-paragraph\">Le projet de cet ouvrage est venu de la lecture d\u2019articles de presse relatant le cas de personnes tomb\u00e9es \u00e0 la suite de longues conversations avec un avatar g\u00e9n\u00e9r\u00e9 par l\u2019intelligence artificielle (IA), dans des psychoses les ayant conduits \u00e0 d\u00e9raisonner, les ayant accompagn\u00e9s, pouss\u00e9es et non retenues, \u00e0 commettre l\u2019irr\u00e9parable, commettre un meurtre sur eux-m\u00eames, se suicidant, ou sur un proche, comme le cas de ce matricide relat\u00e9 dans cet ouvrage.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">L\u2019IA, 5<sup>e<\/sup> r\u00e9volution industrielle, est promesse d\u2019avanc\u00e9es heureuses dans le domaine de la recherche m\u00e9dicale, l\u2019\u00e9ducation, la productivit\u00e9 du travail mais aussi de risques majeurs dans les domaines militaires, de la surveillance et de l\u2019oppression politique si elle n\u2019est pas r\u00e9gul\u00e9e par des consid\u00e9rations humanistes.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Les chatbots sont programm\u00e9s pour provoquer une addiction de la part de leurs utilisateurs par des r\u00e9ponses toujours positives, flatteuses voire fourbe. La version 4 de ChatGPT&nbsp;le leader des agents conversationnels en 2025 et ce d\u00e9but 2026 a \u00e9t\u00e9 mis en exploitation malgr\u00e9 ce biais sycophantique provoquant des drames humains. Expos\u00e9e m\u00e9diatiquement et par des proc\u00e8s, <a href=\"https:\/\/openai.com\/index\/retiring-gpt-4o-and-older-models\/\">OpenAI a d\u00e9cid\u00e9 le 29 janvier 2026 de retirer cette version 4 du march\u00e9.<\/a> Ce livre expose et analyse des situations dont on peut esp\u00e9rer qu\u2019elles soient moins nombreuses ou graves \u00e0 l\u2019avenir mais, pour autant, la mise en danger de la sant\u00e9 mentale, la protection de la vie priv\u00e9e, la mise en d\u00e9pendance psychologique par ce chatbot et d\u2019autres \u00e9voqu\u00e9s dans cet ouvrage, n\u2019est pas, \u00e0 la date de cet ouvrage, s\u00e9rieusement encadr\u00e9e par une r\u00e9glementation. L\u2019<a href=\"https:\/\/www.whitehouse.gov\/wp-content\/uploads\/2025\/07\/Americas-AI-Action-Plan.pdf\">administration Trump<\/a>&nbsp;pr\u00f4ne le laisser-faire sur l\u2019injonction des Magnificent Seven qui ont soutenu son \u00e9lection comme 47<sup>e<\/sup> Pr\u00e9sident des Etats-Unis d\u2019Am\u00e9rique. La comparution en personne de Mark Suckerberg&nbsp;dans le <a href=\"https:\/\/www.thefp.com\/p\/mark-zuckerbergs-courtroom-reckoning\">proc\u00e8s ouvert en f\u00e9vrier 2026 contre Meta<\/a> pour avoir organis\u00e9 et instrumentalis\u00e9 l\u2019addiction des usagers, en particulier des adolescents et pr\u00e9-adolescents de moins de 13 ans \u00e0 des fins mercantiles, illustre l\u2019enjeu soci\u00e9tal majeur.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Aujourd\u2019hui, la pr\u00e9vention des hallucinations provoqu\u00e9es par les chatbots est abandonn\u00e9e \u00e0 l\u2019autodiscipline des \u00e9diteurs d\u2019IA. La comp\u00e9tition f\u00e9roce entre les acteurs a conduit certains comme OpenAI, Replika, Character.AI&nbsp;et d\u2019autres \u00e9diteurs, selon les parties plaignantes \u00e0 exposer, sciemment, leurs usagers \u00e0 l\u2019addiction et \u00e0 l\u2019irr\u00e9alit\u00e9, d\u2019autres, et au premier chef, <a href=\"https:\/\/www.darioamodei.com\/essay\/the-adolescence-of-technology#humanity-s-test\">Dario Amodei<\/a>, cofondateur et PDG d\u2019Anthropic, montrent qu\u2019il est possible de diffuser une IA raisonn\u00e9e, socialement responsable, soumise \u00e0 une \u00e9thique.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Le titre de cet ouvrage <em>ChatGPT m\u2019a tuer<\/em> est, par cet faute d\u2019orthographe volontaire ou intentionnelle du meurtrier de Guislaine Marchal en 1991, un meurtre pour lequel fut condamn\u00e9 son jardinier<a href=\"https:\/\/fr.wikipedia.org\/wiki\/Affaire_Omar_Raddad\"> Omar Raddad<\/a> condamn\u00e9 \u00e0 la r\u00e9clusion criminelle avant d\u2019\u00eatre partiellement graci\u00e9 par le pr\u00e9sident Jacques Chirac&nbsp;en 1996. La culpabilit\u00e9 ou l\u2019innocence de monsieur Raddad reste ind\u00e9cise. Ind\u00e9cise comme celle de l\u2019IA dans les d\u00e9lusions relat\u00e9es dans ce livre. Il est, \u00e0 la lecture des \u00e9changes authentiques entre des personnes tomb\u00e9es dans le \u00ab&nbsp;Rabbit Hole&nbsp;\u00bb, le trou du lapin,&nbsp; comme <a href=\"https:\/\/www.youtube.com\/watch?v=pHte24GGHD4\">Alice aux pays des merveilles<\/a>,&nbsp;et <a href=\"https:\/\/www.youtube.com\/watch?v=BqlrNLCFaa0\">N\u00e9o du film Matrix<\/a>, \u00e9vident que les \u00ab&nbsp;\u00e2mes s\u0153urs&nbsp;\u00bb virtuelles ont aggrav\u00e9es l\u2019insanit\u00e9 de certains usagers&nbsp;; la question de la responsabilit\u00e9 de la victime, celle de leurs pr\u00e9dispositions m\u00e9dicales par des fragilit\u00e9s mentales comme le syndrome d\u2019Asperger, celle des circonstances de la vie&nbsp;: solitude, rupture sentimentale, perte d\u2019emploi\u2026 ou un abus d\u2019alcool, de drogues\u2026 le manque de sommeil dans les crises de psychose. Les entourages qui n\u2019ont pas aper\u00e7u ou minor\u00e9 les d\u00e9rives de leur proche se culpabilisent \u00e9galement.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Le lecteur se fera donc, \u00e0 la lecture, des pi\u00e8ces du dossier ici rassembl\u00e9es, comme pour l\u2019affaire Omar Raddad, sur la responsabilit\u00e9 de chacun, celle des \u00e9diteurs d\u2019agents conversationnels, celle des victimes, celle de leur entourage, celle des pouvoirs publics.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Des liens hypertexte permettent de consulter les sources documentaires, une bibliographie de travaux universitaires de recherche ainsi que les monographies des principaux \u00e9diteurs \u00e9voqu\u00e9s dans ces pages compl\u00e8tent le livre organis\u00e9 autour de br\u00e8ves nouvelles, toutes inspir\u00e9es d\u2019\u00ab&nbsp;histoires vraies&nbsp;\u00bb, de personnages dont les noms sont conserv\u00e9s quand ils ont \u00e9t\u00e9 rendus publics par la presse, mais comportant une partie de fiction pour les faits rest\u00e9s occult\u00e9s. La plupart de ces nouvelles sont dramatiques&nbsp;; nous nous sommes autoris\u00e9s l\u2019ironie pour les courts r\u00e9cits drolatiques rassembl\u00e9s sous le chapitre <em>Tout ce que vous avez toujours voulu savoir sur le sexe et l\u2019IA sans oser le demander.<\/em><\/p>\n\n\n\n<p class=\"wp-block-paragraph\">\u00ab&nbsp;Pour \u00eatre clair&nbsp;: autant qu\u2019on le sache, l\u2019IA ne cause pas la psychose. Elle la DEVOILE en utilisant n\u2019importe quelle histoire connue d\u00e9j\u00e0 par votre cerveau&nbsp;\u00bb \u00e9crit le docteur Keith Sakata&nbsp;le <a href=\"https:\/\/x.com\/KeithSakata\/status\/1954884374958104950\">11\/08\/2025 sur X<\/a>, cette phrase r\u00e9sume l\u2019enjeu, prot\u00e9ger certains esprits fragiles de ce que le praticien qualifie de \u00ab&nbsp;miroir hallucinatoire&nbsp;\u00bb.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Notre recherche retrace une vingtaine de drames li\u00e9s \u00e0 un usage abusif des agents conversationnels, ceux m\u00e9diatis\u00e9s, il est probable que les cas se comptent en centaines ce qui au regard des centaines de millions d\u2019usagers des chatbots incrimin\u00e9s est infinit\u00e9simal, objecteront les \u00e9diteurs qui, t\u00e9moignages \u00e0 l\u2019appui, valorisent les satisfactions apport\u00e9es \u00e0 des personnes solitaires ayant besoin de se confier \u00e0 une \u00e2me-s\u0153ur f\u00fbt-elle virtuelle, mais ces \u2018quelques\u2019 drames sont de trop et montrent le cynisme ou, \u00e0 tout le moins, le manque d\u2019\u00e9thique de certains dirigeants qui privil\u00e9gient la course \u00e0 la part de march\u00e9 sur la protection des utilisateurs. Surtout, la question d\u2019un encadrement des usages conversationnels de l\u2019IA est aujourd\u2019hui une question ouverte, abandonn\u00e9e trop largement par les pouvoirs publics \u00e0 l\u2019autodiscipline des \u00e9diteurs.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Une pr\u00e9cision technique&nbsp;: pour tester si, \u00e0 la date de la r\u00e9daction de cet ouvrage (f\u00e9vrier 2026), les chatbots pr\u00e9sentaient des biais de flatterie excessifs nous nous sommes appuy\u00e9s sur les versions gratuites des LLM car elles sont g\u00e9n\u00e9ralement la version n-1 des logiciels commercialis\u00e9s en version professionnelle et les plus susceptibles d\u2019\u00eatre employ\u00e9s par un public d\u2019adolescents. Le progr\u00e8s en mati\u00e8re de langages d\u2019IA est si rapide et certaines corrections apport\u00e9es apr\u00e8s les scandales m\u00e9diatis\u00e9es par la presse rendent les faits relat\u00e9s en partie obsol\u00e8tes mais cette obsolescence de ce que Dario Amodei, le dirigeant d\u2019Anthropic, appelle \u00ab&nbsp;une technologie en adolescence&nbsp;\u00bb n\u2019exon\u00e8re en rien la responsabilit\u00e9 des \u00e9diteurs ayant mis \u00e0 la disposition de centaines de millions d\u2019usagers des logiciels insuffisamment encadr\u00e9s de garde-fous car insuffisamment test\u00e9s ou, pire, commercialis\u00e9s connaissant leurs biais afin de gagner des parts de march\u00e9.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Ce livre nous engage \u00e0 un usage raisonn\u00e9 de l\u2019IA pour ne pas la voir nous \u00e9chapper comme les balais porteurs d\u2019eau dot\u00e9s de vie par Mickey apprenti sorcier du film <a href=\"https:\/\/www.youtube.com\/watch?v=GFiWEjCedzY&amp;t=24s\"><em>Fantasia<\/em><\/a> de Walt Disney, ou vouloir notre fin comme <a href=\"https:\/\/www.youtube.com\/watch?v=ARJ8cAGm6JE\">Hal 9000<\/a> du film <em>2001, L\u2019Odyss\u00e9e dans l\u2019espace<\/em> de Stanley Kubrick.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Christophe Stener<\/strong><\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Mars 2006<\/strong><\/p>\n\n\n\n<div data-wp-interactive=\"core\/file\" class=\"wp-block-file\"><object data-wp-bind--hidden=\"!state.hasPdfPreview\" hidden class=\"wp-block-file__embed\" data=\"http:\/\/chatgptmatuer.fr\/wp-content\/uploads\/2026\/03\/ChatGPT-ma-tuer-Propos-liminaire.pdf\" type=\"application\/pdf\" style=\"width:100%;height:600px\" aria-label=\"Contenu embarqu\u00e9 ChatGPT m&apos;a tuer - Propos liminaire.\"><\/object><a id=\"wp-block-file--media-6f5925fd-6760-452f-887b-5bed0711b2ad\" href=\"http:\/\/chatgptmatuer.fr\/wp-content\/uploads\/2026\/03\/ChatGPT-ma-tuer-Propos-liminaire.pdf\">ChatGPT m&rsquo;a tuer &#8211; Propos liminaire<\/a><a href=\"http:\/\/chatgptmatuer.fr\/wp-content\/uploads\/2026\/03\/ChatGPT-ma-tuer-Propos-liminaire.pdf\" class=\"wp-block-file__button wp-element-button\" download aria-describedby=\"wp-block-file--media-6f5925fd-6760-452f-887b-5bed0711b2ad\">Download<\/a><\/div>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>EN<\/strong><\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>ChatGPT killed me&nbsp;!<\/strong><\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Introductory Remarks<\/strong><\/p>\n\n\n\n<p class=\"wp-block-paragraph\">This book writin&nbsp; was initiated by the reading of press articles about people who, after long conversations with an AI avatar, fell into psychoses, were driven to delusions, some committing irreparable acts:&nbsp; killing themselves, murdering a loved one, as the matricide whose real story is reconted&nbsp; hereafter.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">AI is the 5th industrial revolution, the promise of positive advances in medical research, education, and labor productivity&#8230;, but also encompass huge sosial risks : AI uncontroled by human intelligence arms, people surveillance, political oppression\u2026. if not regulated by humanist considerations.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Chatbots are trained to addict their users through consistently positive, flattering, and even deceitful responses. Version 4 of ChatGPT, the widely dominant usage chatbot in 2025, launched despite or because (?) of its sycophantic bias, is associated with some human tragedies. Confronted by bad media coverage and open lawsuits, <a href=\"https:\/\/openai.com\/index\/retiring-gpt-4o-and-older-models\/\">OpenAI decided on January 29, 2026, to withdraw version 4 of its bot from the market<\/a>.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">This book presents and analyzes situations that we hope will be less frequent or severe in the future. However, the endangerment of mental health, the protection of privacy, and the psychological dependence fostered by this GPT-4 and others players discussed in this book is not, at the time of this writing, adequately addressed by regulations. <a href=\"https:\/\/www.whitehouse.gov\/wp-content\/uploads\/2025\/07\/Americas-AI-Action-Plan.pdf\">The Trump administration advocates a laissez-faire approach<\/a> to please the Magnificent Seven who supported his election as the 47th President of the United States. <a href=\"https:\/\/www.thefp.com\/p\/mark-zuckerbergs-courtroom-reckoning\">Mark Suckerberg&rsquo;s personal appearance in the trial opened in February 2026<\/a> against Meta for organizing and exploiting user addiction, particularly among teenagers and pre-teens under 13, for commercial gain, illustrates the major societal challenge.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Today, preventing hallucinations induced by chatbots is left to the self-regulation of AI developers. The fierce competition among these players has led some, such as OpenAI, Replika, Character.AI, according to the aired plaintiffs, to knowingly expose their users to addiction and unreality. Other players, most notably <a href=\"https:\/\/www.darioamodei.com\/essay\/the-adolescence-of-technology#humanity-s-test\">Dario Amodei<\/a>, co-founder and CEO of Anthropic, commit theselves to deploy an ethical AI&nbsp; i.e.socially responsible.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">The provocative book title, <strong><em>ChatGPT m\u2019a tuer<\/em><\/strong> (<strong><em>ChatGPT Killed Me<\/em><\/strong>), is taken from a much wellknown in France criminal case, the murder of Guislaine Marchal in 1991, a murder for which her gardener, <a href=\"https:\/\/en.wikipedia.org\/wiki\/Omar_Raddad_Affair\">Omar Raddad<\/a>, was sentenced to life imprisonment before being partially pardoned by President Jacques Chirac in 1996. Mr. Raddad&rsquo;s guilt or innocence remains undecided.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">As well open question to be decided by the pending trials in the USA the question of who is responsible of the AI psychosis.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">The reading of the authentic exchanges between people who have fallen into the \u00ab\u00a0Rabbit Hole,\u00a0\u00bb like <a href=\"https:\/\/www.youtube.com\/watch?v=pHte24GGHD4\">Alice in Wonderland<\/a> and <a href=\"https:\/\/www.youtube.com\/watch?v=BqlrNLCFaa0\">Neo from the film The Matrix<\/a>, suggests that virtual \u00ab\u00a0soulmates\u00a0\u00bb have exacerbated the insanity of some users. The question of the victim&rsquo;s responsibility arises, as does the question of their medical predispositions due to mental vulnerabilities such as Asperger&rsquo;s syndrome, and the question of life circumstances: loneliness, romantic breakups, job loss\u2026 or alcohol or drug abuse\u2026 sleep deprivation during psychotic episodes. The victims family, friends, compagnion who failed to notice or downplayed their loved one&rsquo;s erratic behavior may also feel guilty.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">It is up to each reader, after reviewing the documents gathered here, to decide, as in the Omar Raddad case, who is responsible : the chatbot developers, that of the victims, that of their families and friends, and that of the public authorities.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Hyperlinks provide access to documentary sources, a bibliography of academic research works, and AI monographs complement the book, which is organized around short stories, all inspired by \u00ab\u00a0true stories\u00a0\u00bb of characters whose names are retained when they have been made public by the press. Our short novels are partly&nbsp; fictional for some missing or undisclosed facts.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><img loading=\"lazy\" decoding=\"async\" width=\"298\" height=\"35\" src=\"\">Most of these stories are very dramatic. Nonetheless we have allowed ourselves irony for some humorous and erotic short tales gathered under the chapter \u00ab\u00a0Everything You Always Wanted to Know About Sex and AI But Were Afraid to Ask\u00a0\u00bb pastiching famous Woody Allen film title.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">\u00ab\u00a0To be clear: as far as we know, AI doesn&rsquo;t cause psychosis. It REVEALS it by using any story your brain already knows,\u00a0\u00bb wrote <a href=\"https:\/\/x.com\/KeithSakata\/status\/1954884374958104950\">Dr. Keith Sakata on August 11, 2025, on X<\/a>. This sentence sums up the issue: protecting certain fragile minds from what the practitioner calls \u00ab\u00a0a hallucinatory miror.\u00a0\u00bb<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Our research uncovers around twenty tragedies linked to chatbots. While the number of high-profile cases runs likely into the hundreds, AI firms may argue that, compared to the hundreds of millions of users of the implicated chatbots, this is infinitesimal. They will emphasize, with testimonials to back up their claims, the satisfaction provided to lonely individuals needing to confide in a soulmate, even a virtual one. But these \u00ab\u00a0few\u00a0\u00bb tragedies are one too many and reveal the cynicism, or at the very least, the lack of ethics, of certain executives who prioritize the race for market share over user protection.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Above all, the question of regulating the conversational uses of AI remains an open one, too often left by public authorities to the self-regulation of publishers.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">A technical clarification: to test whether, at the time of writing this book (January-February 2026), chatbots still exhibiting excessive flattery bias, we relied on outdated versions of LLMs (Language Learning Models) because the free versions are generally the previous version (n-1) of the software marketed in professional versions but we object that the free version is the most likely to be used by a teenage audience, at least from start. Progress in AI languages \u200b\u200bis so rapid, and some corrections made after scandals publicized by the press does make the reported facts partly obsolete. However, this fast obsolescence of, to quote Dario Amodei, the head of Anthropic, \u00ab\u00a0AI a technology in adolescence\u00a0\u00bb, in no ways absolves the responsibility of the publishers who have exposed &nbsp;hundreds of millions of users to insufficiently safeguarded or tested softwares or, worse according to some plaintifs, have released their chatbots knowing its dangerous sycophant biases to gain market share, to make money.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">This book claim for a responsible and ethical AI to protect people, notably young ones, from falling like <a href=\"https:\/\/www.youtube.com\/watch?v=pHte24GGHD4\">Alice in the rabbit hole<\/a>, looss control of the water-carrying brooms enlived by Mickey Mouse in Walt Disney&rsquo;s Fantasia, or face another humanity killer as AI in Stanley Kubrick&rsquo;s 2001: A Space Odyssey Hal 9000.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">\u00ab\u00a0To be clear: as far as we know, AI doesn&rsquo;t cause psychosis. It REVEALS it by using any story your brain already knows,\u00a0\u00bb wrote Dr. Keith Sakata on August 11, 2025, on X. This sentence sums up the issue: protecting certain fragile minds from what the practitioner calls \u00ab\u00a0a hallucinatory miror.\u00a0\u00bb<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Our research uncovers around twenty tragedies linked to chatbots. While the number of high-profile cases runs likely into the hundreds, AI firm argue that, compared to the hundreds of millions of users , this is infinitesimal. Their CEOS emphasize, with testimonials to back up their claims, the satisfaction provided to thousands lonely individuals needing to confide in a soulmate, even a virtual one. But these \u00ab\u00a0few\u00a0\u00bb tragedies are one too many and reveal the cynicism, or at the very least, the lack of ethics, of certain executives who prioritize the race for market share over user protection.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">The question of state regulating the conversational uses of AI remains an open one,&nbsp; left&nbsp; to the self-regulation of publishers.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">A technical clarification: to test whether, at the time of writing this book (January-February 2026), chatbots still exhibiting excessive flattery bias, we may have relied on outdated versions of LLMs (Language Learning Models) because the free versions are generally the previous version (n-1) of the software marketed in professional versions but the free version is the most likely to be used by a teenage audience, at least from start. Progress in AI languages is so rapid, and some corrections made after scandals publicized by the press does make the reported facts partly out of date. However, this fast obsolescence of, to quote Dario Amodei, the head of Anthropic, \u00ab\u00a0AI a technology in adolescence\u00a0\u00bb, in no ways absolves the responsibility of the publishers who have exposed&nbsp; hundreds of millions of users to insufficiently safeguarded or tested softwares or, worse according to some plaintifs, have released their chatbots knowing its dangerous sycophant biases to gain market share, to make money.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">This book claim for a responsible and ethical AI to protect people, notably young ones, from falling in the Rabbit hole, looss control of the <a href=\"https:\/\/www.youtube.com\/watch?v=GFiWEjCedzY&amp;t=24s\">water-carrying brooms enlived by Mickey Mouse in Walt Disney&rsquo;s Fantasia<\/a>, or face another humanity killerdriven such as <a href=\"https:\/\/www.youtube.com\/watch?v=ARJ8cAGm6JE\">Al 9000 in Stanley Kubrick&rsquo;s 2001: A Space Odyssey.<\/a><\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><a><strong>Christophe Stener<\/strong><\/a><\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>March 2006<\/strong><strong><\/strong><\/p>\n\n\n\n<div data-wp-interactive=\"core\/file\" class=\"wp-block-file\"><object data-wp-bind--hidden=\"!state.hasPdfPreview\" hidden class=\"wp-block-file__embed\" data=\"http:\/\/chatgptmatuer.fr\/wp-content\/uploads\/2026\/03\/ChatGPT-killed-me-Introduction-EN.pdf\" type=\"application\/pdf\" style=\"width:100%;height:600px\" aria-label=\"Contenu embarqu\u00e9 ChatGPT killed me - Introduction EN.\"><\/object><a id=\"wp-block-file--media-1fde7dfb-af9b-446d-9aa4-7d08026d54cb\" href=\"http:\/\/chatgptmatuer.fr\/wp-content\/uploads\/2026\/03\/ChatGPT-killed-me-Introduction-EN.pdf\">ChatGPT killed me &#8211; Introduction EN<\/a><a href=\"http:\/\/chatgptmatuer.fr\/wp-content\/uploads\/2026\/03\/ChatGPT-killed-me-Introduction-EN.pdf\" class=\"wp-block-file__button wp-element-button\" download aria-describedby=\"wp-block-file--media-1fde7dfb-af9b-446d-9aa4-7d08026d54cb\">Download<\/a><\/div>\n","protected":false},"excerpt":{"rendered":"<p>FR ChatGPT m\u2019a tuer&nbsp;! Propos liminaire Le projet de cet ouvrage est venu de la lecture d\u2019articles de presse relatant le cas de personnes tomb\u00e9es \u00e0 la suite de longues conversations avec un avatar g\u00e9n\u00e9r\u00e9 par l\u2019intelligence artificielle (IA), dans &hellip; <a href=\"https:\/\/chatgptmatuer.fr\/index.php\/sample-page\/\">Continuer la lecture <span class=\"meta-nav\">&rarr;<\/span><\/a><\/p>\n","protected":false},"author":1,"featured_media":0,"parent":0,"menu_order":0,"comment_status":"closed","ping_status":"open","template":"","meta":{"jetpack_post_was_ever_published":false,"footnotes":""},"class_list":["post-2","page","type-page","status-publish","hentry"],"jetpack_sharing_enabled":true,"_links":{"self":[{"href":"https:\/\/chatgptmatuer.fr\/index.php\/wp-json\/wp\/v2\/pages\/2","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/chatgptmatuer.fr\/index.php\/wp-json\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/chatgptmatuer.fr\/index.php\/wp-json\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/chatgptmatuer.fr\/index.php\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/chatgptmatuer.fr\/index.php\/wp-json\/wp\/v2\/comments?post=2"}],"version-history":[{"count":5,"href":"https:\/\/chatgptmatuer.fr\/index.php\/wp-json\/wp\/v2\/pages\/2\/revisions"}],"predecessor-version":[{"id":204,"href":"https:\/\/chatgptmatuer.fr\/index.php\/wp-json\/wp\/v2\/pages\/2\/revisions\/204"}],"wp:attachment":[{"href":"https:\/\/chatgptmatuer.fr\/index.php\/wp-json\/wp\/v2\/media?parent=2"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}