Microsoft's AI chief defends use of open web content
Mustafa Suleyman, the head of AI at Microsoft, has publicly expressed his belief that content available on the open web is akin to "freeware" that anyone can freely utilize. Microsoft is currently the target of multiple lawsuits that accuse the company of using copyrighted online stories, to train its generative AI models. Despite facing legal challenges, Suleyman, in a conversation with CNBC's Andrew Ross Sorkin, defended the practice of using online content to train AI models as perfectly legal.
Advocacy for knowledge sharing amid legal challenges
Suleyman further elaborated on his perspective of humanity's role in knowledge production, stating, "What are we, collectively, as an organism of humans, other than a knowledge and intellectual production engine?" This statement underscores his belief in the open sharing and utilization of web-based information.
Suleyman addresses gray area of robots.txt files
Suleyman also discussed the contentious issue of robots.txt files, which specify which bots are prohibited from scraping a specific website. He acknowledged this as a gray area and predicted it would be addressed in court. Despite being viewed as a social contract since the 1990s, robots.txt is not legally binding and some AI companies, including Microsoft's partner OpenAI, reportedly disregard it.
