A burgeoning field of artificial intelligence research is focused on creating AI systems capable of improving themselves—and other AI—without direct human intervention. Ricursive Intelligence, a startup founded by two former Google researchers, is at the forefront of this effort.

Valued at $4 billion, Ricursive Intelligence is pursuing what’s known as recursive intelligence, a concept where AI systems are designed to automate the process of creating and refining other AI models. This approach aims to accelerate the development of AI and potentially unlock capabilities beyond what human engineers can currently achieve.

The company is not alone in this pursuit. Several other research groups and companies are exploring similar avenues to automate AI creation, recognizing the potential to overcome limitations in current AI development methods. Recursive intelligence promises to address challenges like the time and resources required to train complex AI models, and the need for specialized expertise in AI engineering.

The founders, both previously employed at Google, bring significant experience in machine learning and AI research to Ricursive Intelligence. Their expertise is driving the company’s efforts to build foundational technologies that will enable recursive AI systems. While the technology is still in its early stages, the potential impact across various industries is substantial, ranging from automated software development to advancements in scientific discovery.