Meta and IBM assembled the Avengers of open-sourced artificial intelligence on Tuesday to form the AI Alliance, taking on leaders in the space such as OpenAI, Google, and Anthropic. The group of 50 companies, startups, universities, research institutes, and government bodies aims to foster an open-sourced community and accelerate responsible innovation in AI.
“This is a pivotal moment in defining the future of AI,” said IBM CEO Arvind Krishna in a press release. “IBM is proud to partner with like-minded organizations through the AI Alliance to ensure this open ecosystem drives an innovative AI agenda underpinned by safety, accountability, and scientific rigor.”
If the mission of the AI Alliance sounds familiar to the original mission of OpenAI -– industry and academia team up to create open, responsible AI – that’s because they’re nearly identical. A key difference is the structure, however, because these companies will remain independent, as a network of partners instead of a nonprofit that ends up closed source with a $90 billion valuation. OpenAI’s instability in November proved it may be too unstable to lead the world into the AI revolution. The alliance pools resources from leading institutions to offer a competitive alternative, giving them a leg up as businesses and governments look for partners in shaping the future of generative AI.
Llama2, Stable Diffusion, StarCoder, and Bloom are the largest open-sourced models within the AI alliance. Other notable AI organizations in the alliance are Hugging Face, Intel, AMD, Oracle, and Stability AI. Yale, Cornell, Hebrew University of Jerusalem, and UC Berkeley are some of the educational institutions in the partnership.
The alliance plans to advance the ecosystem of open foundation models, ones that are free to use, as well as develop evaluation tools to enable the responsible development of AI. The coalition also hopes to foster a vibrant AI hardware accelerator by “boosting contributions,” which could mean the open-sourced models on the list will receive a computing chip boost. Lastly, the group seeks to educate the public and regulators about the risks and benefits of AI.