Preeti
OpenAI and Google have announced new collaborations with major publishers to expand paid access to information used by their AI products. OpenAI has partnered with Time Magazine to access the publisher’s assets dating back to 1923, aiming to enhance its AI tools and support reputable journalism. The integration of Time’s content into OpenAI’s products will likely involve the use of retrieval augmented generation (RAG) to generate more accurate answers to user queries.
Google has also expanded its content partnerships with companies like Moody’s, MSCI, Thomson Reuters, and Zoominfo to feed its RAG databases. The goal is to improve the quality of AI-generated results by grounding apps in factual information from these sites. However, the quality of the answers still depends on the quality of the source data, as seen with Google’s AI search function presenting troll posts as fact.
These collaborations come amid a growing demand for high-quality datasets in machine learning tools. Reddit has taken steps to protect its content from being scraped by AI startups without compensation, a practice that OpenAI and Google have agreed to avoid. These deals are lucrative, with Google reportedly paying Reddit $60 million annually for access to its posts.
Despite the potential benefits of these partnerships, challenges remain, such as broken URLs in ChatGPT responses and legal issues surrounding the use of copyrighted content. The Center for Investigative Reporting is suing OpenAI and Microsoft for allegedly using its content without permission or compensation.
Source link
Source link: https://www.theregister.com/2024/06/28/openai_google_ai/
GIPHY App Key not set. Please check settings