AMI Labs News Tracker

LeCun gave a guest lecture at Princeton's Al-Khwarizmi Applied Mathematics Webinar

Yann LeCun argued that today’s AI systems and large language models are fundamentally limited because they lack true world understanding, advocating instead for “world models” that can predict the consequences of actions and ultimately enable safer, more human-like intelligence.

Yann LeCun in conversation with Jacob Effron, partner at Partner at Redpoint

Yann LeCun speaking with WelchLabs.

This was the central question of the debate between Professor Yann LeCun, Turing Award laureate, Executive Chairman of Advanced

Yann LeCun, a leading AI researcher, discusses how current machine learning developments represent the construction of a prefrontal cortex for machines, highlighting advances in artificial intelligence architecture and cognitive capabilities.

World models are an emerging area of artificial intelligence that has recently gained significant attention in the field. These AI systems are designed to develop internal representations of how the world works, enabling them to understand and predict the dynamics of their environment. World models represent a promising development in machine learning that could enhance AI's ability to reason, plan, and interact with complex systems.

Goldman Sachs is working on developing a "world model" for artificial intelligence, addressing what is considered a critical missing component in current AI systems. The world model represents a key technological challenge that major AI researchers and companies are actively pursuing to advance machine intelligence capabilities.

Quentin Duval is an accomplished Artificial Intelligence Research Engineer specializing in large-scale machine learning systems, with a strong focus on self-supervised learning for images and video. Currently a Member of Technical Staff at AMI and formerly at Meta’s Facebook AI Research lab in Montréal, he has been a core contributor to state-of-the-art video generation models such as Emu-Video and MovieGen. His work emphasizes optimizing large-scale training architectures, including advanced parallelism techniques, as well as building robust data pipelines and evaluation systems for high-performance AI models.

Chao Du is a Senior Research Scientist specializing in machine learning and deep learning, with a strong track record of advancing AI research across both academia and industry. Currently based in Singapore at Sea AI Lab (SAIL), he leads and contributes to cutting-edge research initiatives that push the boundaries of scalable, applied artificial intelligence.

John Nguyen is a Member of Technical Staff at AMI Labs, where he focuses on building visually intelligent systems and advancing research toward unified world models. His work sits at the intersection of multimodal generation, perception, and understanding, with a particular emphasis on open-world intelligence and large-scale system design.

Christophe Tcheng is a seasoned engineering leader with extensive experience building and scaling complex technology platforms across global organizations. Currently serving as VP of Engineering at AMI – Advanced Machine Intelligence, he brings deep expertise in software development, organizational leadership, and product platform strategy.

Jeff Bezos is assembling a specialized artificial intelligence team for Project Prometheus, his initiative to develop advanced machine intelligence capabilities. The effort appears to be conducted with minimal public visibility, suggesting a focus on building technological expertise in AI development away from mainstream attention.

As our first on-the-ground US recruiter, you'll own AMI's talent pipeline for our New York and Montreal offices — with the flexibility to support hiring across our global offices as needed.

A $1B wager on world models signals a radical shift away from LLMs toward AI systems that learn causality from the real world.

Matthew J. Muckley has joined AMI Labs as a Member of the Technical Staff, where he will focus on advancing physical world models, an area central to AMI’s broader mission of building robust, generalizable AI systems.

After launching three AI startups, Alexandre Lebrun has embarked on the ‘project of a lifetime’ as CEO of LeCun’s new startup, Advanced Machine Intelligence.

AI pioneer Yann LeCun delivered a lecture at Brown University discussing emerging developments and new frontiers in artificial intelligence. The presentation, covered by The Brown Daily Herald, featured insights from one of the field's most influential researchers on the current state and future directions of AI technology.

Yann LeCun, a prominent AI researcher, delivered a lecture at Brown University discussing a novel approach to artificial intelligence. The talk likely covered LeCun's latest perspectives on advancing machine learning and AI methodologies.

Jihan Yang, a leading emerging voice in multimodal artificial intelligence, is advancing the frontier of machine learning as a postdoctoral associate at New York University’s Courant Institute of Mathematical Sciences.

Xinyi Wan, a seasoned machine learning researcher and systems engineer, has joined AMI Labs as a Member of Research Staff, bringing with him over a decade of experience at the intersection of large-scale infrastructure and applied artificial intelligence.

A leading AI researcher with publications at top conferences (CVPR, ICCV, ECCV, ICLR) and significant contributions at FAIR—including V-JEPA 2 and widely adopted open-source models—has joined AMI Labs to advance human-like visual understanding in machines. At AMI, he brings deep expertise in self-supervised learning, multimodal AI, and large-scale applied systems to push the frontier of accessible and intelligent AI.

As Nabla gains international attention through its close ties to Yann LeCun’s AMI, one of the most ambitious AI research labs, Soffiati joins at a pivotal moment in the shift toward next-generation, agentic healthcare systems.

A breakthrough in world models: LeCun’s team shows how simpler, smaller AI systems can learn to predict reality and still be faster, cheaper, and without the fragile tricks that once made JEPAs so hard to train.

LeWorldModel is a new JEPA-based world model that learns stable latent representations directly from pixels using just two loss terms. It avoids collapse without complex tricks, improves efficiency and speed, and enables meaningful physical reasoning in latent space.

AMI Labs co-founders Saining Xie sits for a 7-hour marathon interview.

AMI Labs has posted a new HR Ops position within the General and Administration department, based in Paris. The role reflects the company's continued organisational build-out following its $1 billion fundraise in early 2026. Applications can be submitted directly via the AMI Labs Ashby jobs board.

A designer explores a critical gap in AI world models related to how systems handle iconic or symbolic representations, highlighting a blind spot in current advanced machine intelligence systems. The article presents field observations on how AI systems struggle to process and understand iconic visual or conceptual elements compared to other types of data. This research contributes to understanding limitations in AI perception and reasoning that designers and developers should account for when building intelligent systems.

AMI Labs explores new approaches to redefining intelligence, challenging conventional frameworks in machine intelligence development. The initiative suggests a shift in how advanced machine intelligence systems are conceptualized and evaluated.

Yann LeCun's AMI Labs (Advanced Machine Intelligence) has raised $1 billion in seed funding, marking a significant investment in AI development. The funding round suggests growing momentum in the field of world models, an emerging area of AI research focused on creating systems that can understand and simulate the physical world.

AMI Labs raised $1 billion in funding while positioning itself against the generative AI trend that dominates the broader AI industry. The startup's contrarian approach suggests a focus on alternative AI methodologies rather than the large language models and generative systems that have become mainstream.

AMI Labs (Advanced Machine Intelligence) was featured in healthcare AI news coverage on March 11, 2026. The article discusses developments in AI applications within the healthcare industry. Specific details about AMI Labs' announcements or initiatives would be found in the full article content.

Yann LeCun has secured $1 billion in funding to develop AI technology focused on understanding the physical world. This initiative aims to advance machine intelligence capabilities beyond current limitations by creating systems that can better comprehend and interact with physical environments.

Advanced Machine Intelligence (AMI) has announced a $1.03 billion funding round. The substantial investment round represents significant capital injection for the company's development and expansion efforts.

AMI Labs, a start-up founded by a former Meta A.I. chief, has reached a valuation of $3.5 billion, according to a report by The New York Times. The company's significant valuation reflects investor confidence in its artificial intelligence capabilities and leadership.

Advanced Machine Intelligence (AMI) is advancing the next phase of artificial intelligence development through the creation of foundational world models. These world models represent a key technological breakthrough that enables more sophisticated AI capabilities. The innovation positions AMI Labs at the forefront of next-generation AI research and development.

AMI Labs, led by Polytechnique alumnus Alexandre Lebrun, has secured $1 billion in funding to advance artificial intelligence technology. The company aims to significantly transform the AI landscape with this substantial investment.

Yann LeCun, a former Meta AI pioneer, has raised over $1 billion in funding for his new startup, AMI Labs (Advanced Machine Intelligence), according to reporting by the Wall Street Journal. The significant capital raise underscores investor confidence in LeCun's vision for advancing machine intelligence technology outside of major tech companies.

Yann LeCun's AI startup AMI Labs has secured a $1 billion seed round funding, backed by major investors including Nvidia and Temasek. The substantial investment reflects significant confidence in the company's advanced machine intelligence technology and potential in the AI sector.

Yann LeCun has raised $1 billion in funding to establish AMI Labs (Advanced Machine Intelligence), focused on developing AI systems with improved understanding of physical world dynamics and reasoning. The funding reflects growing investment in foundational AI research aimed at moving beyond current language models toward more physically-grounded artificial intelligence.

# Summary AMI Labs, a startup founded by renowned AI researcher Yann LeCun, has raised $1.03 billion in funding to develop world models—AI systems designed to understand and predict how the physical world works. The company aims to advance machine intelligence by training these sophisticated models.

# Summary AMI Labs (Advanced Machine Intelligence), founded by renowned AI researcher Yann LeCun, has secured a record $1 billion seed funding round. The Paris-based startup's exceptional early-stage capital raises questions about the competitive landscape in AI development and venture funding priorities.

Yann LeCun has secured $1 billion in funding for AMI Labs to advance research into world models, positioning this approach as a correction to current directions in AI development. The funding reflects LeCun's argument that the AI industry has fundamentally misunderstood key aspects of artificial intelligence advancement.

# Summary Yann LeCun, a veteran from Meta, has secured $1 billion in funding for his AI startup AMI Labs (Advanced Machine Intelligence). The substantial investment underscores growing investor confidence in LeCun's artificial intelligence venture and his vision for advancing machine intelligence technology.

# AMI Labs Secures $1 Billion Funding with New CEO Yann LeCun's AI startup, AMI Labs (Advanced Machine Intelligence), has announced a new CEO and secured $1 billion in funding. The leadership change and capital raise mark a significant milestone for the company as it advances its machine intelligence initiatives.

# Summary A prominent AI researcher dubbed the "AI godfather" has raised $1.3 billion in funding for a new start-up that will establish Singapore as a key operational base. The funding round reflects significant investor confidence in the company's AI technology and Singapore's growing importance as a hub for artificial intelligence development.

# AMI Labs Funding Summary AMI Labs, founded by former Meta AI chief Yann LeCun, has raised $1.03 billion in funding to develop an alternative approach to artificial intelligence. The significant investment underscores LeCun's vision for a different direction in AI development outside of his previous role at Meta.

# AMI Labs Funding Summary AMI Labs, founded by Yann LeCun, has secured $1.03 billion in funding to advance research and development of world models—AI systems designed to understand and predict how the physical world works. The funding represents a significant investment in next-generation artificial intelligence technology focused on building more sophisticated machine intelligence capabilities.

# AMI Labs Funding Summary AMI Labs, founded by former Meta AI chief Yann LeCun, has raised $1.03 billion in funding to develop an alternative approach to artificial intelligence. The funding round reflects investor confidence in LeCun's vision for a different direction in AI development compared to current industry approaches.

French AI startup AMI (Advanced Machine Intelligence) has announced raising over $1 billion in seed funding, making it one of the largest seed rounds in European startup history. Founded by Yann LeCun alongside a team of former Meta FAIR researchers, AMI is developing 'universal intelligent systems' based on world models. The company is headquartered in Paris with offices in New York, Montreal, and Singapore.

Channel News Asia reports that AMI Labs, led by former Meta Chief AI Scientist Yann LeCun, has raised $1.03 billion to advance its alternative approach to AI development centred on world models and the JEPA architecture, positioning itself as a counterpoint to mainstream large language model-based AI systems.

Toyota Ventures has published a blog post explaining its strategic investment in AMI Labs, noting that the company's physical AI and world model research directly aligns with Toyota's vision for next-generation automotive and robotics intelligence. Toyota Ventures sees AMI's JEPA architecture as a promising foundation for AI systems that must reason about the physical world.

AMI Labs, founded by Yann LeCun, has secured $1.03 billion in funding to develop world models, representing a significant investment in advanced machine intelligence research. The substantial funding round underscores growing interest in foundation models and AI systems capable of understanding and predicting complex real-world environments.

AMI Labs (Advanced Machine Intelligence), the AI startup co-founded by Turing Award laureate Yann LeCun, has raised $1.03 billion in a seed round at a $3.5 billion valuation. The company is building world models based on LeCun's JEPA (Joint Embedding Predictive Architecture) framework, which aims to create AI systems that understand the physical world. The round was co-led by Bezos Expeditions, Cathay Innovation, Greycroft, Hiro Capital, and HV Capital.

Hiro Capital has announced co-leading a $1.03 billion seed round in Advanced Machine Intelligence (AMI Labs) alongside Bezos Expeditions, Cathay Innovation, Greycroft, and HV Capital. AMI is focused on Physical AI and world models — areas that align closely with Hiro Capital's investment thesis around spatial computing and physical intelligence.

AMI Labs, the AI startup led by Yann LeCun — who served as Meta's Chief AI Scientist for over a decade — has been valued at $3.5 billion after closing a $1.03 billion seed round. The funding underscores continued investor appetite for foundational AI research, particularly approaches that diverge from the dominant large language model paradigm championed by OpenAI and Google.

AMI Labs has secured over $1 billion in seed funding, with notable angel investors including Jeff Bezos, Eric Schmidt, Mark Cuban, Xavier Niel, Tim Berners-Lee, and Jim Breyer participating alongside institutional investors. The company's approach centers on JEPA-based world models as an alternative path to artificial general intelligence, with a 13-person founding team drawn predominantly from Meta's Fundamental AI Research (FAIR) lab.

With $1.03B in seed funding, Paris-based AMI launches as a $3.5B AI unicorn aiming to move beyond LLMs by building “world models. These systems are designed to understand physical reality, reason about cause and effect, and power the next generation of intelligent machines.

Ex-Meta Chief AI Scientist Yann LeCun pushes back on AGI hype, arguing AI is far from human-level intelligence and better understood as a tool to amplify people. He says the real breakthrough lies not in language models, but in building world models that can navigate the messy physical world.

# Summary Yann LeCun, executive chairman of AMI Labs, stated that current AI technology has significant limitations and is primarily effective at language-based tasks. According to LeCun, existing AI systems lack broader capabilities beyond language processing, suggesting the field needs further advancement to achieve more comprehensive artificial intelligence.

Yann LeCun has launched AMI Labs, a new venture that takes a contrarian approach by moving away from the large language model paradigm that currently dominates AI development. The company's strategy represents a deliberate pivot against the prevailing trend of scaling up language models as the primary path forward in machine intelligence.

AMI filed for its fifth trademark with France’s National Institute of Intellectual Property (INPI).

AMI filed for its fourth trademark with France’s National Institute of Intellectual Property (INPI).

AMI filed for its third trademark with France’s National Institute of Intellectual Property (INPI).

AMI filed for its second trademark with France’s National Institute of Intellectual Property (INPI).

Yann LeCun, a prominent AI researcher, has launched a new startup called AMI Labs (Advanced Machine Intelligence) focused on world models technology. The Forbes article examines whether AMI Labs can help advance world models beyond their current hype and prove their practical value in AI development.

PengyouCo updated its articles of association on 19 December 2025, restructuring its 1,000,000 shares: ordinary shares were increased from 33,500 to 71,750, while SP preference shares (which carry 10× voting rights) were adjusted to 928,250. The update was certified by founding president Gérard Lebrun and preceded the company's January 2026 fundraise announcement under the AMI Labs brand.

Yann LeCun is raising €500 million in funding for his new AI startup AMI Labs at a €3 billion valuation. The funding round reflects significant investor confidence in LeCun's advanced machine intelligence venture.

The legendary AI scientist left Meta to build world models. He says Silicon Valley is too hypnotized by ChatGPT to see what's coming next.

PengyouCo — the legal entity that became AMI Labs (Advanced Machine Intelligence) — was incorporated on 3 December 2025 as a Société par Actions Simplifiée (SAS) with €10,000 share capital. The company was registered at 51 rue de Seine, Paris 6e, with Gérard Lebrun (father of CEO Alexandre LeBrun) as sole founding shareholder and first president. Its stated corporate purpose covers AI model development, software engineering, and R&D.

AMI filed for its first trademark with France’s National Institute of Intellectual Property (INPI).

``` AMI Labs has released Janus Pro, an open-source image generation model. The model was presented on Hacker News as a novel advancement in machine intelligence for image synthesis. ```