Perplexity logo.

The British public broadcaster BBC has announced legal action against the American artificial intelligence (AI) startup perplexity, marking the beginning of a copyright conflict between media organizations and generative AI corporations.

On the 20th (local time), BBC sent an official letter to perplexity Chief Executive Officer Aravind Srinivasan, pointing out that the company's AI model had used BBC article content without permission. Crawling refers to the practice of taking web pages as they are and extracting data from them. BBC stated that it suggested halting the use of its content immediately, deleting all content in its possession, and providing monetary compensation for already used content.

BBC claimed that its article content was replicated word for word in some of perplexity's AI responses, and that about 17% of those responses contained factual inaccuracies or distortion of context. As a public broadcaster, BBC emphasized that continued unauthorized use would not only damage its credibility but also threaten its license-based financial operations.

In response, perplexity stated, "BBC lacks understanding of technology, the internet, and intellectual property rights," and rebutted the claims as "manipulative and opportunistic." It further clarified that it does not develop or train base models independently, but is merely an interface platform that accesses various external models based on OpenAI, Google, Anthropic, and Meta.

Perplexity is also involved in legal disputes with Dow Jones, a subsidiary of News Corporation, as well as with the New York Post and The Wall Street Journal. It has also received cease and desist letters from other media organizations, including The New York Times and Condé Nast.

This case has emerged as a symbolic event regarding copyright issues in the AI industry, being reported by numerous major international media outlets such as Reuters, Financial Times, and The Guardian. There is a significant possibility that this issue will extend beyond a simple corporate dispute to an international discussion on how the generative AI industry should handle media copyrights. The industry is suggesting that not only could there be controversies over the use of training data, but also that changes in AI-based search and news consumption methods could fundamentally shake the existing business models of media organizations.

※ This article has been translated by AI. Share your feedback here.