Join our daily and weekly newsletters for the latest updates and exclusive content on industry-leading AI coverage. Learn more
François Chollet, a former Google engineer and the creator of the widely used Python deep learning framework It's difficultis co-founded gooda new AI research and science lab, with Mike Knoop, co-founder of Zapier.
In a post on the new startup websitethe founders explained their goals in combining intuitive pattern recognition, enabled by deep learning, with formal reasoning through what they called “guided program synthesis.”
According to them, this integration will allow AI systems to adapt and innovate beyond current task-specific applications, ultimately leading to artificial general intelligence (AGI), which loosely which is defined throughout the AI community as machine intelligence that can outperform humans in the most economically important way. and cognitive tasks. As they write:
“We need computers that can pose problems and explore new territory, not just apply known solutions. We need computers that can innovate. The path to AGI is not through incremental which is an improvement over existing methods.”
The duo is yet to say whether or not they have received outside funding for this venture or are bootstrapping it with their own funds.
This comes months after OpenAI's former co-founder and chief scientist Ilya Sutskever, who reportedly led the briefly successful but ultimately backfired internal coup against his fellow co-founder, Sam Altman, announced also a startup focused on developing “Safe Superintelligence” with $1 billion in private backing.
Beyond deep learning
While current deep learning systems are impressive, Chollet and Knoop argue that they are primarily limited by their reliance on large datasets and their inability to adapt efficiently to new tasks.
Chollet and Knoop believe that program synthesis is the key to overcoming these limitations.
Unlike traditional deep learning, which interpolates between data points, program synthesis looks for discrete programs that explain the data. This method allows for broader generalizations with fewer data points.
Combining the intuitive capabilities of deep learning with the rigorous reasoning of program synthesis could lead to a new paradigm for AI research.
“Ndea's mission is to operate AGI to realize unprecedented scientific progress for the benefit of all present and future generations,” they said.
Building a “Factory for Scientific Development”
Ndea's long-term vision goes beyond creating AGI. The lab aims to act as a “factory for rapid scientific advancement,” capable of solving both known and unknown challenges.
From tackling current frontiers like autonomous vehicles and sustainable energy to accelerating entirely new discoveries, the lab sees itself as a catalyst for scientific progress.
Chollet added that their research direction has the potential to open breakthroughs and redefine the boundaries of human knowledge. As he wrote in a thread on X: “If we are successful, we will not stop at AI. With this technology in hand, we want to tackle every scientific problem that it can solve. We see the acceleration of scientific progress as the most exciting application of AI.”
According to Chollet, this progress depends on developing AI that can learn as well as humans and continuously improve over time without bottlenecks.
While acknowledging that success is not guaranteed, Chollet emphasized the importance of accomplishing this ambitious goal, telling X: “We believe we have a small but real chance of achieving a breakthrough—creating the AI that can continue to improve over time with no apparent bottlenecks.”
A New Research Focus for AGI
Program synthesis, the foundation of Ndea's research, is still a relatively young field. Chollet likened its current situation to that of an in-depth study in 2012.
However, he noted that its potential is increasingly being recognized by frontier AI labs, even though most see it as only a small part of what is needed for AGI.
Ndea, by contrast, considers program synthesis as important as deep learning and has made it central to their approach.
The lab is also actively recruiting a globally distributed team of researchers and engineers to form what it describes as the most “talent-dense program synthesis team” in the world.
The company operates as a fully remote organization and seeks candidates with strong technical expertise, particularly in translating mathematical concepts into code.
Founders with a strong track record
François Chollet and Mike Knoop bring extensive experience to Ndea.
At Google, Chollet worked on fundamental research in deep learning and AI systems, gaining insights into the limitations of existing models and opportunities for improvement. His contributions include not only Keras but also ARC-AGI benchmark, a widely used metric for measuring progress toward AGI.
He is also the author of the book Deep Learning with Python and recognized in Time's “100 Most Influential People in AI.”
Knoop founded Zapier, the world's largest AI automation company, where he led engineering and product development as well as the company's early adoption of AI technologies.
He is also credited with pioneering best practices for globally distributed teams. Both Chollet and Knoop are co-founders of the ARC Prize Foundation, a nonprofit organization dedicated to advancing open AGI research.
Visions of the future are rooted in ancient tradition
Ndea gets its name from Greek concepts ennoia (intuitive understanding) and dianoia (logical reasoning), which reflects the lab's goal of combining deep learning and program synthesis. By running AGI, Ndea hopes to compress centuries of scientific progress into decades or even years.
While acknowledging the uncertainty and challenges of pursuing AGI, Chollet and Knoop remain optimistic about their approach. They see AGI as the gateway to addressing humanity's toughest challenges and uncovering new opportunities for discovery.
