Meta Platforms and International Business Machines (IBM) have jointly unveiled the AI Alliance, a coalition comprised of over 50 artificial intelligence companies and research institutions.
The AI alliance aims to promote an "open innovation and open science" model within the AI sector, featuring key contributors like Intel, Oracle, Cornell University, and the National Science Foundation.
Darío Gil, IBM's Senior Vice President and Director of IBM Research disclosed that the formation of the AI Alliance stemmed from dissatisfaction with the perceived lack of diversity in recent AI discussions. The alliance emphasizes an open-source approach, aligning with the historical collaboration among major tech firms, academic institutions, and independent programmers.
The spotlight on generative AI, particularly since the introduction of OpenAI's ChatGPT a year ago, has intensified the narrative. OpenAI and counterparts like Anthropic and Cohere have been at the forefront of developing proprietary AI systems, whereas the AI Alliance's members, including industry players such as Intel and Oracle, are striving to carve out a space in the rapidly expanding AI market.
IBM, despite past challenges with its Watson system, is introducing the Watsonx system as a novel platform. Meta, while endeavoring to bolster its presence in the AI market, is advocating for its Llama 2 AI model as an open-source AI system.
"This other way, it's a much more distributed approach, but much more resilient, because no given institution can derail the success of the open engine," Gil said, as quoted by The Wall Street Journal.
Mitigating Risks
Against the backdrop of disruptions at OpenAI in November, businesses are increasingly exploring alternative AI providers to mitigate the risks associated with relying on a single vendor. The AI Alliance serves as a collaborative platform for organizations navigating this shift.
The alliance's formation underscores the intensifying debate on the merits and risks associated with adopting an open-source approach to AI development.
Yann LeCun, Meta's Chief AI Scientist, has critiqued major players, including OpenAI, Google, and Anthropic, for perceived lobbying efforts favoring rules that could consolidate power over AI development, per AP News.
In response, the Frontier Model Forum, comprising Meta, Microsoft, Google, and Anthropic, has been established to address industry concerns.
The Center for Humane Technology, a vocal critic of Meta's social media practices, has emphasized the potential risks tied to open-source or leaked AI models. This ongoing debate raises critical questions about the responsible deployment of AI models to the public.
AI Alliance Open for More Players
IBM's role in the AI Alliance prompts inquiries into its motivations as a co-founder alongside Meta. Despite increased earnings from generative AI, IBM faces stiff competition from Microsoft and OpenAI in the enterprise-focused AI services sector.
Notable exclusions from the AI Alliance's initial membership, such as Stanford, and MIT, and AI startups like Anthropic, Cohere, and Adept, have raised queries. IBM's response suggests a focus on organizations strongly committed to open innovation, leaving room for potential additions to the alliance.
Sriram Raghavan, VP of IBM's research AI division, has stated that the AI Alliance's current focus centers on members strongly committed to open innovation and open-source AI. Anticipating further evolution, he envisions more organizations joining in the future, as reported by TechCrunch.
Related Article : How AI Is Shaping Australia's Job Landscape and Going Unnoticed
