{"id":1665385,"date":"2026-06-22T14:33:01","date_gmt":"2026-06-22T18:33:01","guid":{"rendered":"https:\/\/observer.com\/?p=1665385"},"modified":"2026-06-22T14:33:01","modified_gmt":"2026-06-22T18:33:01","slug":"yann-lecun-startup-ami-labs-future-ai","status":"publish","type":"post","link":"https:\/\/observer.com\/2026\/06\/yann-lecun-startup-ami-labs-future-ai\/","title":{"rendered":"How Yann LeCun\u2019s Startup Challenges the Logic Behind Today\u2019s A.I. Race"},"content":{"rendered":"
\"Yann
The Turing Award winner says A.I. needs broader foundations than just scaling text models. Yui Mok \/ POOL \/ AFP via Getty Images<\/span><\/figcaption><\/figure>\n

Meta\u2019s former chief A.I. scientist, Yann LeCun, helped lay the groundwork for the modern A.I. boom long before chatbots became an industry obsession. Now, after\u00a0leaving Meta<\/span> in late 2025, he is building a multibillion-dollar startup<\/a> under the premise that the race toward A.I. superintelligence is starting from the wrong place. Speaking at the VivaTech conference on June 17, LeCun shared more details<\/span><\/a>\u00a0about why he believes his startup, Advanced Machine Intelligence (AMI Labs), is on a better path toward human-level A.I. Earlier this year, the Paris-based company\u00a0raised $1.03 billion<\/span> at a $3.5 billion valuation<\/a>, giving LeCun and his team fresh backing to pursue an alternative to today\u2019s generative A.I. models developed by tech giants.<\/p>\n

\u201cI think there is a requirement for new paradigms to go beyond the limitations of current systems,\u201d LeCun, 65, told Wired\u2019s Steven Levy onstage.<\/span><\/p>\n

Building better language models, as many A.I. companies are doing, could be one path, LeCun said, but \u201cit\u2019s kind of a slow way.\u201d According to LeCun, AMI Labs is building “world models” that learn from reality, understand consequences, predict what happens next, and choose the best actions based on those predictions. \u201cWe have good hopes that this will really be a complete change of the blueprints of intelligent systems,\u201d LeCun said.<\/span><\/p>\n

LeCun believes much of Silicon Valley has become \u201cLLM-pilled\u201d \u2014 or overly convinced that scaling large language models will eventually produce human-like intelligence. The result, he argued, is a \u201cmonoculture\u201d of A.I. systems that are powerful in some areas but still fundamentally limited.<\/p>\n

While LLMs are already highly capable in areas like coding, they are built around language. Real-world data from cameras, robots, and other sensors is messier and harder to predict, which is why he believes a different approach is necessary. For AMI Labs, the result could be \u201cconsiderably more intelligent systems that can anticipate the outcome of their own actions, can plan, can reason and perhaps all the way to human intelligence,\u201d he said.<\/span><\/p>\n

Coming from LeCun, the critiques carry weight. A longtime New York University professor, he is among the trio of researchers often referred to as the \u201cGodfathers of A.I.\u201d The others are Geoffrey Hinton, a University of Toronto professor and former Googler, and Yoshua Bengio, a University of Montreal professor who founded Quebec\u2019s A.I. institute, Mila.<\/a> In 2018, they\u00a0collectively won<\/span><\/a> the prestigious Turing Award for breakthroughs that helped push deep learning into the mainstream.<\/p>\n

From FAIR to AMI Labs<\/b><\/h3>\n

In 2013, LeCun\u00a0joined Facebook<\/span><\/a> to launch what would become Facebook A.I. Research, or FAIR, and later transitioned to chief A.I. scientist at Meta. He said AMI has its roots as an internal FAIR project and had support from company leaders, including Mark Zuckerberg and CTO Andrew Bosworth. But as Meta\u2019s A.I. priorities shifted toward catching up with competitors, the environment became less favorable.<\/p>\n

At that point, it made sense to \u201cjust leave the company, try to accelerate the development, and build the things we can build with it,\u201d LeCun said. \u201cAnd I realized I could also raise enough money to keep this going.\u201d<\/span><\/p>\n

In March, AMI Labs raised a $1.03 billion seed round\u2014the largest in European startup history. Investors include Nvidia, Jeff Bezos and Eric Schmidt, among others. With LeCun serving as executive chairman, the startup is led by CEO Alex LeBrun, a former FAIR engineering head and co-founder of health A.I. company Nabla, AMI Labs\u2019 first partner. The company has assembled a research-heavy team with experience at Meta, Google DeepMind and other leading A.I. labs.<\/p>\n

AMI says it aims to build applications in areas where this approach matters, including industrial process control, automation, wearable devices, robotics, and health care. \u201cIt\u2019s not going to happen next year,\u201d LeCun said. \u201cWe\u2019re not going to have a country of geniuses in a data center next year, okay?\u201d<\/p>\n

At VivaTech, the discussion also broadened to a bigger question: who should control the next generation of A.I.? If future assistants are built only by \u201ca handful of companies\u201d in Silicon Valley or China, LeCun warned, \u201cculture is in big trouble\u201d and \u201cdemocracy is in big trouble.\u201d Blocking open-source models in the name of safety, he argued, misses the role A.I. can play in spreading knowledge.<\/a><\/p>\n

That stance also informs LeCun\u2019s work as chief science adviser to Project Tapestry, an AI Alliance-led effort to build a shared global A.I. model through a distributed network of contributors. The AI Alliance is a nonprofit group, co-launched in 2023 by Meta and IBM, to support open-source A.I. development. With Tapestry, the goal is to let countries, companies, universities and others contribute data or computing power without giving up control of that data. As A.I. begins to mediate what people read, watch, and search for, LeCun argues that people need alternatives to what a few dominant American or Chinese companies decide to build.<\/p>\n

LeCun\u2019s comments also come as Meta\u2019s own open-source stance has grown more complicated. Llama 2, released in 2023 for free commercial use, made Meta a rare Big Tech counterweight to more closed A.I. rivals. But Zuckerberg has since signaled<\/span><\/a>\u00a0that Meta may not release all of its future \u201csuperintelligence\u201d models openly.<\/p>\n

\u201cWe need access to a wide diversity of A.I. assistants for the same reason we need access to a wide diversity of the press to get multiple sources of information,\u201d LeCun said. \u201cThe only way I can see that this can happen is if there is an open, free foundation model, on top of which anybody can build their own specialized assistant for their language or languages, their culture, their value system, their political biases, their centers of interest. And so open source has to exist.\u201d<\/span><\/p>\n","protected":false},"excerpt":{"rendered":"

Yann LeCun explains why his startup AMI Labs is pursuing world models, not just larger language models, in a bid to reach human-level A.I.<\/p>\n Read More<\/a>","protected":false},"author":129757851,"featured_media":1665386,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"apple_news_api_created_at":"2026-06-22T18:33:13Z","apple_news_api_id":"00e808a0-be42-4a76-aeb8-dabefa2df458","apple_news_api_modified_at":"2026-06-22T18:33:14Z","apple_news_api_revision":"AAAAAAAAAAD\/\/\/\/\/\/\/\/\/\/w==","apple_news_api_share_url":"https:\/\/apple.news\/AAOgIoL5CSnauuNq--i30WA","apple_news_cover_media_provider":"image","apple_news_coverimage":0,"apple_news_coverimage_caption":"","apple_news_cover_video_id":0,"apple_news_cover_video_url":"","apple_news_cover_embedwebvideo_url":"","apple_news_is_hidden":"","apple_news_is_paid":"","apple_news_is_preview":"","apple_news_is_sponsored":"","apple_news_maturity_rating":"","apple_news_metadata":"\"\"","apple_news_pullquote":"","apple_news_pullquote_position":"","apple_news_slug":"","apple_news_sections":[],"apple_news_suppress_video_url":false,"apple_news_use_image_component":false,"_jetpack_memberships_contains_paid_content":false,"footnotes":"","jetpack_publicize_message":"","jetpack_publicize_feature_enabled":true,"jetpack_social_post_already_shared":true,"jetpack_social_options":{"image_generator_settings":{"template":"highway","default_image_id":0,"font":"","enabled":false},"version":2},"_wpas_customize_per_network":false},"post_tag":[423992964],"company":[171],"channel":[186,12374,423875666],"location":[],"nyo_column":[],"person":[423895646,341024,423938893,423897320,423897335,423904797,596113,423884107,424010230],"nyo_post_hidden":[],"coauthor":[424014214],"class_list":{"0":"post-1665385","1":"post","2":"type-post","3":"status-publish","4":"format-standard","5":"has-post-thumbnail","7":"tag-turing-award","8":"observer_company-meta","9":"channel-business","10":"channel-artificial-intelligence","11":"channel-technology","12":"nyo_person-yann-lecun","13":"nyo_person-mark-zuckerberg","14":"nyo_person-steven-levy","15":"nyo_person-geoffrey-hinton","16":"nyo_person-yoshua-bengio","17":"nyo_person-andrew-bosworth","18":"nyo_person-jeff-bezos","19":"nyo_person-eric-schmidt","20":"nyo_person-alex-lebrun"},"acf":{"homepage_position":"","homepage_title":"","homepage_excerpt":"","alternative_og_image":"","headline":{"seo_headline":""},"subheadline":{"optimized_seo_description":"","optimized_social_excerpt":""}},"apple_news_notices":[],"parsely":{"version":"1.1.0","canonical_url":"https:\/\/observer.com\/2026\/06\/yann-lecun-startup-ami-labs-future-ai\/","smart_links":{"inbound":0,"outbound":0},"traffic_boost_suggestions_count":0,"meta":[],"rendered":"","tracker_url":"https:\/\/cdn.parsely.com\/keys\/observer.com\/p.js"},"jetpack_publicize_connections":[],"jetpack_featured_media_url":"https:\/\/observer.com\/wp-content\/uploads\/sites\/2\/2026\/06\/GettyImages-2244695301.jpg?quality=80","coauthors_byline":"By Beth Duckett","display_channel":"","thumbnail":"\"Yann