Industry stalwarts IBM, Intel, Sony, and Dell have joined forces with Meta, several renowned universities, and a cadre of AI startups and foundations to announce the establishment of the “AI Alliance.” This alliance signals a collaborative effort to disrupt the perceived dominance of established players such as OpenAI, Microsoft, Google, and Amazon in the ever-evolving field of artificial intelligence (AI).
The announcement comes at a time when there have been discussions about the notable absence of traditional computing leaders in the forefront of AI conversations. Dario Gil, a senior vice president at IBM and head of the company’s research lab, expressed concern about the narrow focus on a handful of institutions in the past year’s discourse on AI, emphasizing, “The reality is that this field is much, much larger than that.”
The AI Alliance aligns with a longstanding debate within the developer community regarding the merits of “open” versus “closed” development of AI. Despite its name, OpenAI, primarily backed by Microsoft, has kept its AI algorithms tightly controlled, permitting access only with permission. Similarly, other leading AI developers like Google and Amazon have not fully open-sourced their models, citing competition and safety concerns.
This proprietary approach has sparked debates within the research community and among competing businesses. The formation of the AI Alliance signals a departure from this closed model, with over 50 participants rallying around shared goals.
The objectives of the AI Alliance include establishing common frameworks for evaluating AI algorithm strength, allocating capital to AI research funds, and fostering collaboration on open-source models. This diverse group comprises corporate giants, chip manufacturers such as AMD and Cerebras, AI startups like Hugging Face and Stability AI, and esteemed universities like Yale, Cornell, and Dartmouth.
Darío Gil highlighted examples of the alliance’s commitment to openness, citing IBM’s collaboration with NASA on an open-sourced AI model trained on geospatial data. This model is designed to track deforestation and predict crop yields. Gil also revealed IBM’s substantial financial commitment of approximately $100 million to support AI research projects at universities over the next five years. Additionally, IBM collaborated with Meta to develop an open-source toolkit for AI development.
Regarding governance, Gil mentioned that the alliance is still finalizing details, focusing initially on defining objectives and building a coalition. The next steps involve establishing “technical working groups” for the diverse participants and designing a governance structure, potentially leading to the creation of an external nonprofit. The emergence of the AI Alliance sets the stage for a collaborative and open approach to AI development, challenging the status quo and fostering innovation across the industry.