Meta Adds Reuters Content To AI Chatbot

Meta Platforms said it has formed a deal with Reuters to use the news service’s content in its artificial intelligence (AI) chatbot to answer users’ questions in real time about news and current events.

The deal comes at a time when Meta has been reducing the accessibility of news through some of its platforms in some markets, amdist pressure from governments and online publishers over revenue-sharing and misinformation.

The deal is the first Meta has reached with a news company in several years, and its first related to AI.

Reuters news content became accessible via Meta’s chatbot, known as Meta AI, on Friday, the company said.

News queries

“Through Meta’s partnership with Reuters, Meta AI can respond to news-related questions with summaries and links to Reuters content,” Meta said.

“This partnership will help ensure a more useful experience for those seeking information on current events.”

“We can confirm that Reuters has partnered with tech providers to license our trusted, fact-based news content to power their AI platforms,” Reuters said in a statement, addign that the terms of such deals remains confidential.

The chatbot is integrated into the search and messaging features on Facebook, Instagram WhatsApp and Messenger.

Answers to users’ questions will cite Reuters content and will link to its articles.

The news service is compensated under the deal, but terms of the deal remain confidential, according to an earlier Axios report.

Content controversy

The companies didn’t disclose whether the arrangement involved licensing Reuters content to train Meta’s large language model, Llama.

AI companies are under pressure to ensure their offerings don’t spread misinformation, in a year in which multiple elections are taking place around the world.

OpenAI, the developer of ChatGPT, has made a number of deals with dozens of national, international and local media firms, while OpenAI investor Microsoft said it would pay news companies to place their content in its Copilot AI chatbot.

AI companies including OpenAI, Microsoft and Jeff Bezos-backed startup Perplexity have also been at the receiving end of lawsuits from news media companies alleging they made use of copyrighted content to train their language models.