IBM and Meta Team Up to Support Open AI Projects

IBM and Meta collaborate to boost open AI projects, fostering innovation and shared advancements

IBM and Meta have collaboratively established an AI Alliance, unveiling the initiative with over 50 founding members, including Anyscale, AMD, Cerebras, Cornell University, Dartmouth College, Dell, Stability AI, and Oracle. The alliance, aiming for open collaboration, strives to expedite responsible AI innovation while prioritizing scientific rigor, trust, safety, security, and economic competitiveness. Despite the absence of OpenAI, Microsoft, Google, and AWS, the AI Alliance underscores its commitment to fostering a community focused on open AI advancements. The launch aligns with the industry's growing interest in open-source tools, emphasizing safety and trust as primary objectives. The Safety and Trust Alliance working group has already commenced its meetings, laying the groundwork for the alliance's overarching goals.

IBM's Director of AI Open Innovation, Anthony Annunziata, expressed the belief that open innovation, science, and research are integral to ensuring AI's societal benefits and safety. Emphasizing the need for openness, the AI Alliance plans to establish working groups with a specific focus on safety, trust, and the creation of foundational tools and technologies. Annunziata highlighted the intention to meet regularly, identify priorities, and assess existing tools within the safety and trust alliance working group. To ensure effectiveness, additional working groups, including public policy and advocacy, will be established, along with a steering committee overseeing the entire program.

Futurum Group analyst Mark Beccue commended the AI Alliance's working group approach, emphasizing its efficacy in achieving tangible outcomes. While acknowledging the benefits of open source, Beccue noted its potential downsides, such as security concerns and the assumption that enterprises possess the expertise to navigate open-source technologies. He highlighted the likelihood of enterprises adopting a hybrid approach, combining open-source and closed-source tools to meet their specific needs.

The AI Alliance's launch coincides with a year marked by AI innovation growth, notably the widespread adoption of generative AI and increased interest in open-source tools like Meta's Llama 2. The alliance's focus on safety, trust, and responsible innovation addresses the challenges and opportunities presented by open-source technologies. Meta's absence from leading generative AI alliances led to speculation about its motivations, with analysts suggesting that Meta may leverage the AI Alliance to differentiate itself from other industry leaders.

Chirag Shah, a professor at the University of Washington Information School, highlighted the potential benefits of the AI Alliance in encouraging oversight of open-source AI tools while offering Meta a platform to showcase its distinctiveness. He noted that the alliance could serve as a strategy for Meta and IBM to maintain relevance in a market likely to be dominated by a few major players.