BlockBeats News, June 20th, according to the Financial Times, the British Broadcasting Corporation (BBC) has taken action for the first time against a tech company suspected of copyright infringement, threatening legal action against the artificial intelligence search engine startup Perplexity. BBC stated that there is evidence that Perplexity's "default AI model" was trained using its content. BBC mentioned that unless Perplexity stops scraping content and proposes an "economic compensation scheme" to remedy the intellectual property infringement, it may seek a court injunction.

Perplexity responded, stating that BBC's accusations are "manipulative and opportunistic," and claiming that there is a "fundamental misunderstanding" of technology, the internet, and intellectual property law. "These accusations also demonstrate BBC's willingness to go to any lengths to serve its own interests in order to maintain Google's illegal monopoly status." Perplexity does not build or train base models but provides an interface for users to choose among models developed by other companies such as OpenAI, Google, and Anthropic. According to insiders, Perplexity's internal model is based on Meta's Llama and has been optimized to improve accuracy and reduce instances of the model generating false information.