Big tech companies including Microsoft-backed OpenAI, Google and Google-backed Anthropic are using online content created by other companies to train their generative AI models. However, they do not allow their own data to be used by others to train similar AI models. This has led to a brewing legal battle that could determine the future of the web and how copyright laws work in this new world. Reddit plans to start charging for access to its data, and publishers, including News Corp., are pushing tech companies to pay for the use of their content to train AI models. OpenAI’s CEO, Sam Altman, is working on new AI models that respect copyright. Steven Sinofsky, former Microsoft executive, recently said that the current way AI models are trained breaks the web.
OpenAI is a research laboratory consisting of the for-profit corporation OpenAI LP and its parent company, the non-profit OpenAI Inc. The company’s mission is to ensure that artificial intelligence benefits humanity.
Steven Sinofsky is a former Microsoft executive who is now a board partner at the venture capital firm Andreessen Horowitz.
