What Sets Sakana AI Apart?
Sakana AI, the Japanese startup founded by former Google researchers David Ha and Llion Jones and former diplomat Ren Ito, has recently made headlines by raising $100 million in a series A round of funding. This investment, led by industry heavyweights like New Enterprise Associates, Khosla Ventures, and Lux Capital, with participation from Nvidia, demonstrates the strong traction the year-old company has gained.
Unlike many other AI labs globally, Sakana AI has already established itself as an organization that operates at an impressive speed. The company’s co-founder, David Ha, emphasized this in a tweet, stating that Sakana AI is operating much faster than most frontier AI labs worldwide and has ambitious plans to push the frontiers of what’s possible with AI.
One of the key factors that sets Sakana AI apart is its unique approach to AI development. The company has adopted a nature-inspired collective intelligence approach to building high-performing foundation models. This approach involves bringing together multiple smaller AI models, much like a swarm, to deliver complex results. With this method, Sakana AI has been able to develop various models, including those capable of generating Japan’s traditional ukiyo-e artwork.
Furthermore, Sakana AI recently introduced the AI Scientist, an LLM-based system that automates the entire research lifecycle. This system covers everything from ideation and writing code to running experiments, summarizing results, and even conducting peer-review. By automating these processes, Sakana AI is revolutionizing how research is conducted, making it more efficient and accessible.
To scale up its nature-inspired approach to AI development, Sakana AI plans to use the series A capital and infrastructure support from Nvidia to build a “talent-dense” AI research organization. The company intends to speed up its hiring process and expand its infrastructure within Japan. As part of its partnership with Nvidia, Sakana AI will have access to the latest GPU systems and Nvidia-powered data centers, enabling them to develop advanced models and run experiments more efficiently.
Rivals Eyeing Growth in Japan
While Sakana AI continues to gain traction, it faces competition from other AI labs looking to expand their footprint in Japan. OpenAI, one of the leading research labs, launched a Tokyo hub and released a custom GPT-4 model optimized for the Japanese language. They have also teased a new AI model, GPT-Next, which they claim will be 100x better than the current version.
Additionally, Canada-based enterprise AI startup Cohere is building a custom Japanese model on top of Command R+ in partnership with Fujitsu. These developments highlight the growing interest in Japan’s AI market and the competition among companies to establish a presence in the country.
Despite the competition, investors remain bullish on Sakana AI’s potential. Vinod Khosla, the founder of Khosla Ventures, noted that Sakana AI is showing the “path to innovation” by taking a unique approach to train foundation models. While other labs are using similar techniques, Sakana AI is paving the way for new possibilities in AI development.
In conclusion, Sakana AI’s recent funding round and its unique approach to AI development have positioned it as a key player in the industry. With the support of leading venture capital firms and Nvidia, Sakana AI aims to build a world-class AI lab in Japan. By focusing on nature-inspired collective intelligence and automating the research process, Sakana AI is poised to drive innovation and help Japan regain its technological edge in AI.