IBM and Meta are leading a groundbreaking initiative in the realm of artificial intelligence (AI) by launching the AI Alliance. This new coalition, comprising a mix of corporations, universities, and organizations, aims to promote “open science” and “open technologies.” The alliance includes tech heavyweights like AMD, academia from Cornell, Harvard, and Yale Universities, and organizations such as NASA and Intel.
The AI Alliance has an action-oriented agenda focusing on equitable technology evolution. Its primary goal is to drive responsible innovation by reinforcing trust, safety, and scientific integrity in AI. This will involve developing benchmarks, and standards for evaluation, enhancing global AI skills, and promoting the use of responsible AI.
A yet-to-be-formed governing board and a technical oversight committee will guide the Alliance, emphasizing the importance of collective governance in AI development.
IBM’s SVP Darío Gil highlights the critical juncture AI is approaching – a choice between consolidated control by a few or an open, collaborative approach. The Alliance represents the latter, striving for a more inclusive and safer AI future.
Transparency: The Heart of the AI Debate
The Alliance’s vision resonates with the ongoing debate in AI: closed versus open-source technology. Closed source models, like ChatGPT, Microsoft, and Google’s AI, are restricted to company access only. In contrast, open-source models, such as Meta’s Llama and IBM’s geospatial model, prioritize accessibility and transparency.
Proponents of open-source AI argue that this approach democratizes technology and fosters necessary industry transparency. With closed-source models, in-depth research and understanding of AI capabilities and environmental impacts are challenging.
The Risks and Rewards of Open AI
While open AI offers many advantages, it also comes with risks. Experts like Gary Marcus caution against the potential misuse of open-source AI, especially in creating misinformation and bioweapons. However, advocates like Clément Delangue of Hugging Face assert that open-source AI is essential to balance the power and secrecy of proprietary AI.
The Broader Perspective
The formation of the AI Alliance, though commendable, is just a step towards addressing AI’s broader issues. Experts like Srinivas Mukkamala of Ivanti emphasize the need for more such initiatives to tackle AI governance and equitable access to data. The goal is to avoid a future where AI adoption exacerbates global inequality.
The public’s trust in tech companies self-regulating AI is low, underscoring the need for broader oversight and cooperation. As AI continues to evolve rapidly, initiatives like the AI Alliance represent a significant movement towards a more responsible and inclusive future for AI technology.