At VB Transform 2024, IBM's Cox presented a persuasive argument for open innovation in enterprise artificial intelligence. The speaker, Vice President of machine learning frameworks and director at the MIT-IBM Watson AI Lab, emphasized the importance of openness in technological advancement, particularly in the present AI landscape.
Cox highlighted the nuanced essence of accessibility in AI, contesting the two-dimensional understanding of the principle. He noted the expanding network of open models from various sources, including tech giants, universities, and governments.
Nonetheless, he voiced concerns about the quality of accessibility in various LLMs, noting that some deliver scarcely more than a collection of data points lacking transparency in their creation techniques.