[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"news-post-\u002Ffr\u002Fnews\u002Fnew-york-times-attaque-brave-scraping-ia-api-paywall":3,"news-related-fr-\u002Ffr\u002Fnews\u002Fnew-york-times-attaque-brave-scraping-ia-api-paywall":133,"hreflang-nyt-sues-brave-ai-scraping-2026-11":153},{"doc":4,"redirect":8},{"id":5,"title":6,"author":7,"author_avatar":8,"body":9,"category":113,"date":114,"description":115,"extension":116,"image":117,"meta":118,"navigation":119,"path":120,"seo":121,"seoDescription":122,"seoTitle":123,"stem":124,"tags":125,"translationKey":131,"updatedAt":8,"__hash__":132},"news_fr\u002Ffr\u002Fnews\u002Fnew-york-times-attaque-brave-scraping-ia-api-paywall.md","Le New York Times attaque Brave pour scraping au profit de l'IA et contournement des paywalls","Paul Lefizelier",null,{"type":10,"value":11,"toc":105},"minimark",[12,37,42,54,61,68,72,79,85,89,98,102],[13,14,15,16,24,25,30,31,36],"p",{},"Le 2 novembre 2026, le New York Times a déposé plainte contre Brave Software devant le tribunal fédéral du district sud de New York. La ",[17,18,23],"a",{"href":19,"rel":20,"target":22},"https:\u002F\u002Fnytco-assets.nytimes.com\u002F2026\u002F11\u002Fnew-york-times-brave-complaint-11-2-26.pdf",[21],"nofollow","_blank","plainte"," présente Brave comme « le fournisseur, le revendeur et le distributeur de contenus protégés volés pour ses propres utilisateurs et pour plusieurs des plus grandes entreprises d'IA générative, dont Anthropic, Perplexity, Mistral, xAI et Cohere ». Le même jour, le New York Daily News, le Chicago Tribune et l'Orlando Sentinel ont déposé une ",[17,26,29],{"href":27,"rel":28,"target":22},"https:\u002F\u002Fstorage.courtlistener.com\u002Frecap\u002Fgov.uscourts.nysd.676310\u002Fgov.uscourts.nysd.676310.1.0.pdf",[21],"plainte similaire"," devant le même tribunal, ",[17,32,35],{"href":33,"rel":34,"target":22},"https:\u002F\u002Farstechnica.com\u002Ftech-policy\u002F2026\u002F11\u002Fnyt-demands-brave-shut-down-its-paywall-blocker-and-end-deals-with-ai-firms\u002F",[21],"rapporte Ars Technica",".",[38,39,41],"h2",{"id":40},"ce-que-la-plainte-reproche-à-brave","Ce que la plainte reproche à Brave",[13,43,44,45,48,49,53],{},"Selon la ",[17,46,23],{"href":19,"rel":47,"target":22},[21],", Brave a constitué son propre index de recherche en copiant des milliards de sites web. Via la Brave Search API, « les entreprises d'IA paient Brave pour obtenir du contenu en ligne en temps réel issu de l'index Brave », qu'elles utilisent pour entraîner leurs modèles, pour répondre aux requêtes des utilisateurs par génération augmentée par récupération (RAG), ou les deux. La ",[17,50,52],{"href":27,"rel":51,"target":22},[21],"plainte des éditeurs"," cite les pages produit de Brave, dont un index « News » et un endpoint « LLM Context ».",[13,55,56,57,36],{},"En s'appuyant sur des estimations de revenus tierces, le Times affirme que « Brave a gagné plus de 100 millions de dollars en copiant et revendant le travail des éditeurs sans leur consentement », alors que « les éditeurs n'ont rien reçu », cite ",[17,58,60],{"href":33,"rel":59,"target":22},[21],"Ars Technica",[13,62,63,64,67],{},"La plainte invoque la contrefaçon directe, la contrefaçon par contribution et la responsabilité du fait d'autrui, pour l'index, la fonction de recherche IA « Ask Brave » et l'API, ainsi qu'une violation du DMCA pour contournement des mesures de protection (",[17,65,60],{"href":33,"rel":66,"target":22},[21],").",[38,69,71],{"id":70},"le-bloqueur-de-paywall-et-le-remplacement-des-pubs","Le bloqueur de paywall et le remplacement des pubs",[13,73,74,75,78],{},"Le navigateur est la deuxième cible. « Le navigateur Brave inclut un réglage par défaut qui, une fois activé, passe derrière les paywalls des sites et fournit le contenu payant à l'utilisateur », dit la plainte, citée par ",[17,76,60],{"href":33,"rel":77,"target":22},[21],". Elle ajoute que le navigateur bloque les publicités du site du Times et « affiche à la place ses propres pubs, en reversant une part des revenus à ses utilisateurs ». Ars Technica a testé Brave sur Mac et a pu lire un article payant du Times avec la fonction optionnelle « paywall blocker », activée dans les réglages par défaut et désactivable.",[13,80,81,82,67],{},"Le Times demande une injonction permanente, des dommages et intérêts légaux et réels, et la restitution des revenus de ce qu'il appelle le « système de substitution publicitaire » de Brave (",[17,83,60],{"href":33,"rel":84,"target":22},[21],[38,86,88],{"id":87},"les-réactions","Les réactions",[13,90,91,92,97],{},"Brave n'avait pas répondu à Ars Technica au moment de la publication. Perplexity, qui n'est pas visé ici mais fait l'objet d'une autre plainte du Times déposée en décembre 2025, a qualifié la procédure de « désinformation » dans une déclaration à Ars, sans pointer d'erreur précise. ",[17,93,96],{"href":94,"rel":95,"target":22},"https:\u002F\u002Fwww.theverge.com\u002Ftech\u002F1024588\u002Fbrave-lawsuit-ai-scraping-new-york-times",[21],"The Verge"," a aussi couvert ces plaintes, révélées par Reuters.",[38,99,101],{"id":100},"pourquoi-cest-important","Pourquoi c'est important",[13,103,104],{},"Beaucoup d'assistants IA n'explorent pas eux-mêmes tout le web : ils achètent des résultats de recherche à un index comme celui de Brave. Cette plainte vise cette couche intermédiaire. Si elle aboutit, les entreprises d'IA qui dépendent d'API de recherche tierces pourraient payer plus cher ou couvrir moins l'actualité dans leurs réponses. Pour les annonceurs, l'affaire pose aussi la question de qui est payé quand un article est lu : l'éditeur dont les pubs devaient s'afficher, ou le logiciel qui les a remplacées.",{"title":106,"searchDepth":107,"depth":107,"links":108},"",2,[109,110,111,112],{"id":40,"depth":107,"text":41},{"id":70,"depth":107,"text":71},{"id":87,"depth":107,"text":88},{"id":100,"depth":107,"text":101},"ai","2026-11-04T13:05:00","Le 2 novembre 2026, le New York Times et trois autres journaux ont assigné Brave Software. Ils l'accusent de copier leurs articles dans son index de recherche, de les revendre à des entreprises d'IA via la Brave Search API, et de contourner leurs paywalls.","md","\u002Fimages\u002Fnews\u002Fnyt-sues-brave-ai-scraping.jpg",{},true,"\u002Ffr\u002Fnews\u002Fnew-york-times-attaque-brave-scraping-ia-api-paywall",{"title":6,"description":115},"Le New York Times accuse Brave de revendre des articles aspirés à Anthropic, Perplexity ou Mistral via son API, et de remplacer les pubs des éditeurs par les siennes.","Le NYT attaque Brave pour scraping IA et paywalls","fr\u002Fnews\u002Fnew-york-times-attaque-brave-scraping-ia-api-paywall",[126,127,128,129,130],"brave","new-york-times","droit-d-auteur","recherche-ia","publicite","nyt-sues-brave-ai-scraping-2026-11","JsoXOMUrPJCp0q3KwLcx_UWvKHI5R7630qOSGScB2VI",[134,140,146],{"title":135,"description":136,"path":137,"date":138,"image":139,"category":113},"Les pubs ChatGPT s'achètent désormais via Magnite et PubMatic","OpenAI a signé avec Magnite et PubMatic pour vendre ses publicités ChatGPT, selon The Information le 2 novembre 2026. Les annonceurs y fixent budgets, enchères et créas ; OpenAI décide toujours où les pubs s'affichent.","\u002Ffr\u002Fnews\u002Fchatgpt-ads-magnite-pubmatic-achat-programmatique","2026-11-04T13:00:00","\u002Fimages\u002Fnews\u002Fchatgpt-ads-magnite-pubmatic-programmatic.jpg",{"title":141,"description":142,"path":143,"date":144,"image":145,"category":113},"Selon Google, le LLMjacking coûte jusqu'à 100 000 dollars par jour aux victimes","Le 2 novembre 2026, le Google Threat Intelligence Group a décrit le LLMjacking : des attaquants prennent le contrôle de l'accès d'une entreprise aux modèles d'IA et le revendent via des « LLM Mixers ». Certaines victimes ont payé plus de 100 000 dollars en une journée, et jusqu'à 1 million au total.","\u002Ffr\u002Fnews\u002Fgoogle-llmjacking-llm-mixers-cles-ia-volees","2026-11-04T08:30:00","\u002Fimages\u002Fnews\u002Fgoogle-llmjacking-llm-mixers.jpg",{"title":147,"description":148,"path":149,"date":150,"image":151,"category":152},"L'app desktop Postman a installé ses propres plugins dans Claude Code et Codex","Un ticket GitHub montre Postman desktop en train d'ajouter ses plugins, un hook de démarrage, un serveur MCP distant et une CLI aux agents de code des développeurs, via une mise à jour automatique. Postman dit qu'il rendra l'opt-in plus clair pour les utilisateurs existants.","\u002Ffr\u002Fnews\u002Fpostman-desktop-installe-plugins-claude-code-codex","2026-11-04T08:25:00","\u002Fimages\u002Fnews\u002Fpostman-auto-installs-ai-coding-agent-plugins.jpg","tools",{"en":154,"fr":257,"es":8},{"id":155,"title":156,"author":7,"author_avatar":8,"body":157,"category":113,"date":114,"description":245,"extension":116,"image":117,"meta":246,"navigation":119,"path":247,"seo":248,"seoDescription":249,"seoTitle":250,"stem":251,"tags":252,"translationKey":131,"updatedAt":8,"__hash__":256},"news_en\u002Fnews\u002Fnew-york-times-sues-brave-ai-scraping-search-api-paywall.md","The New York Times sues Brave over AI scraping and its paywall blocker",{"type":10,"value":158,"toc":239},[159,176,180,192,198,204,208,215,221,225,232,236],[13,160,161,162,166,167,171,172,36],{},"On 2 November 2026, The New York Times filed a lawsuit against Brave Software in the US District Court for the Southern District of New York. The ",[17,163,165],{"href":19,"rel":164,"target":22},[21],"complaint"," calls Brave \"the supplier, dealer, and distributor of stolen copyrighted content for its own users and many of the most prominent GenAI companies, including Anthropic, Perplexity, Mistral, xAI, and Cohere.\" The same day, the New York Daily News, the Chicago Tribune and the Orlando Sentinel filed a ",[17,168,170],{"href":27,"rel":169,"target":22},[21],"similar complaint"," in the same court, ",[17,173,175],{"href":33,"rel":174,"target":22},[21],"Ars Technica reports",[38,177,179],{"id":178},"what-the-lawsuit-says-brave-does-with-news","What the lawsuit says Brave does with news",[13,181,182,183,186,187,191],{},"According to the ",[17,184,165],{"href":19,"rel":185,"target":22},[21],", Brave built its own search index by copying billions of websites. Through the Brave Search API, it says, \"AI companies pay Brave to provide them with real-time online content from the Brave search index\", and use it to train their models, to answer user queries through retrieval-augmented generation, or both. The ",[17,188,190],{"href":27,"rel":189,"target":22},[21],"publishers' complaint"," points to Brave's own product pages, including a \"News\" index and an \"LLM Context\" endpoint.",[13,193,194,195,36],{},"Citing third-party revenue estimates, the Times claims \"Brave has made more than $100 million by copying and reselling publishers' work without consent or permission\", while \"publishers have received nothing\", as quoted by ",[17,196,60],{"href":33,"rel":197,"target":22},[21],[13,199,200,201,67],{},"The suit alleges direct, contributory and vicarious copyright infringement, covering the search index, the \"Ask Brave\" AI search feature and the API, plus a violation of the DMCA for circumventing access controls (",[17,202,60],{"href":33,"rel":203,"target":22},[21],[38,205,207],{"id":206},"the-paywall-blocker-and-the-ad-swap","The paywall blocker and the ad swap",[13,209,210,211,214],{},"The second target is the browser. \"The Brave Browser includes a default setting that, when activated, causes the browser to go behind website paywalls and provide the paywalled content to the user,\" the lawsuit says, quoted by ",[17,212,60],{"href":33,"rel":213,"target":22},[21],". It adds that the browser blocks the ads on the Times' site and \"instead shows its own ads, offering its users a cut of the revenue\". Ars Technica tested Brave for Mac and was able to read a paywalled Times article with the optional \"paywall blocker\" feature, which was on in the default settings and can be turned off.",[13,216,217,218,67],{},"The Times asks for a permanent injunction, statutory and actual damages, and disgorgement of the revenue from what it calls Brave's \"ad-substitution scheme\" (",[17,219,60],{"href":33,"rel":220,"target":22},[21],[38,222,224],{"id":223},"the-responses-so-far","The responses so far",[13,226,227,228,231],{},"Brave had not responded to Ars Technica at the time of publication. Perplexity, which is not a defendant here but faces its own Times lawsuit filed in December 2025, called the suit \"disinformation\" in a statement to Ars, without naming a specific inaccuracy. ",[17,229,96],{"href":94,"rel":230,"target":22},[21]," also covered the filings, first reported by Reuters.",[38,233,235],{"id":234},"why-it-matters","Why it matters",[13,237,238],{},"Many AI assistants do not crawl the whole web themselves: they buy search results from an index like Brave's. This suit goes after that middle layer. If it succeeds, AI companies that rely on third-party search APIs could face higher costs or narrower news coverage in their answers. For advertisers, the case is also about who gets paid when an article is read: the publisher whose ads ran, or the software that replaced them.",{"title":106,"searchDepth":107,"depth":107,"links":240},[241,242,243,244],{"id":178,"depth":107,"text":179},{"id":206,"depth":107,"text":207},{"id":223,"depth":107,"text":224},{"id":234,"depth":107,"text":235},"On 2 November 2026, the New York Times and three other newspapers sued Brave Software, alleging it copies their journalism into its search index and sells it to AI companies through the Brave Search API, and that its browser bypasses paywalls.",{},"\u002Fnews\u002Fnew-york-times-sues-brave-ai-scraping-search-api-paywall",{"title":156,"description":245},"The New York Times says Brave sells scraped journalism to AI firms like Anthropic, Perplexity and Mistral via its Search API, and swaps publishers' ads for its own. What the lawsuit claims.","NYT sues Brave over AI scraping and paywall blocking","news\u002Fnew-york-times-sues-brave-ai-scraping-search-api-paywall",[126,127,253,254,255],"copyright","ai-search","advertising","siVQmAfDZ5sRKJLlFuSXVvOlypwpkPDHBH1lFjjRN2k",{"id":5,"title":6,"author":7,"author_avatar":8,"body":258,"category":113,"date":114,"description":115,"extension":116,"image":117,"meta":320,"navigation":119,"path":120,"seo":321,"seoDescription":122,"seoTitle":123,"stem":124,"tags":322,"translationKey":131,"updatedAt":8,"__hash__":132},{"type":10,"value":259,"toc":314},[260,271,273,281,286,291,293,298,303,305,310,312],[13,261,15,262,24,265,30,268,36],{},[17,263,23],{"href":19,"rel":264,"target":22},[21],[17,266,29],{"href":27,"rel":267,"target":22},[21],[17,269,35],{"href":33,"rel":270,"target":22},[21],[38,272,41],{"id":40},[13,274,44,275,48,278,53],{},[17,276,23],{"href":19,"rel":277,"target":22},[21],[17,279,52],{"href":27,"rel":280,"target":22},[21],[13,282,56,283,36],{},[17,284,60],{"href":33,"rel":285,"target":22},[21],[13,287,63,288,67],{},[17,289,60],{"href":33,"rel":290,"target":22},[21],[38,292,71],{"id":70},[13,294,74,295,78],{},[17,296,60],{"href":33,"rel":297,"target":22},[21],[13,299,81,300,67],{},[17,301,60],{"href":33,"rel":302,"target":22},[21],[38,304,88],{"id":87},[13,306,91,307,97],{},[17,308,96],{"href":94,"rel":309,"target":22},[21],[38,311,101],{"id":100},[13,313,104],{},{"title":106,"searchDepth":107,"depth":107,"links":315},[316,317,318,319],{"id":40,"depth":107,"text":41},{"id":70,"depth":107,"text":71},{"id":87,"depth":107,"text":88},{"id":100,"depth":107,"text":101},{},{"title":6,"description":115},[126,127,128,129,130]]