Meta Eyes Paid Partnerships with News Publishers for AI Training Data
### Meta’s Generative AI Gamble: A New Chapter with the News Industry?
Meta, the tech titan formerly known as Facebook, is diving headfirst into the world of generative AI. In its pursuit of excellence, the company is now contemplating the possibility of paying for better quality and more immediate training data. The focus? The news industry. According to insiders, Meta is deliberating internally whether to establish new paid agreements with news publishers to gain deeper access to their content, including news articles, photos, and videos. These considerations stem from Meta’s need to make its generative AI tools, such as Meta AI, more effective and competitive in the ever-crowded arena of AI search tools and chatbots.
Interestingly, Meta has yet to formally approach any news outlets regarding licensing or accessing content. Should they decide to proceed, any resulting agreements would be distinct from prior deals where Meta compensated publishers to host links to their content on its platforms. This shift in strategy marks a significant departure from Meta’s previous engagements with the news industry. As recently as last year, the company slashed a $2 billion budget allocated for its News division, signaling a sharp turn away from its past dealings with news publishers.
Mark Zuckerberg, Meta’s CEO, has previously asserted that the company possesses its own substantial data for training its Llama large language model. This data, he claimed, is larger than Common Crawl, a massive set of scraped web data widely used for AI model training. However, relying solely on its own data might place Meta at a disadvantage, potentially lagging behind competitors like Google and OpenAI. These rivals have already forged deals with news publishers and media outlets to secure more access to content for model training, ensuring their AI tools remain robust and up-to-date.
The landscape for generative AI has evolved rapidly since the launch of ChatGPT almost two years ago, which brought the technology into the broader public consciousness. In response, news outlets and other websites began blocking automated bots deployed by Common Crawl and OpenAI to prevent their content from being scraped for free. Adding to the complexity, the US Copyright Office is currently considering new rules to govern generative AI. Without free and constant access to high-quality news publisher content, Meta AI’s responses to user prompts about current events may become limited, outdated, or even incorrect.
As Meta ponders its next move, it faces a critical decision: whether to invest in paid deals with news publishers to enhance its generative AI capabilities. Major tech companies are already ahead in this race, having secured agreements with news outlets to bolster their AI models. For news publishers, such licensing deals are a pragmatic choice, as they offer some compensation for their valuable content. As one insider put it, “Something is better than nothing.”
Ultimately, Meta’s journey into generative AI and its potential collaborations with the news industry could reshape the landscape of AI-driven content and user experiences. Whether Meta chooses to invest in these partnerships remains to be seen, but one thing is clear: the stakes are high, and the race for AI supremacy is far from over.
