Meta’s new terms may allow it to train AI on news sites for free
Meta’s updated business terms appear to let it crawl websites associated with businesses using its platforms and use their content to develop and train AI products. The change could give Meta access to publishers’ material without separate licensing agreements, raising questions about how online content is used.
The terms take effect on 30 October and apply to businesses promoting content on Facebook, Instagram, WhatsApp and Messenger. They permit automated tools to access and analyse associated web pages and URLs; businesses that disagree may lose access to Meta’s products for commercial use. Meta has licensing deals with CNN, Fox News, USA Today and People Inc., while its new Meta One subscription charges some users to include links in posts and reels.
- Meta’s new terms permit crawling of websites associated with business accounts.
- Content could be used to train AI without a separate publisher licence.
- Businesses that reject the terms may lose commercial platform access.
New here? Start with this
Facebook, Instagram, WhatsApp and Messenger are owned by Meta, which offers businesses tools to promote their products and content on those platforms. Publishers, such as newspapers and broadcasters, often link their websites to social media to reach readers and attract visitors.
AI systems are developed by analysing large amounts of material, including text from websites. Publishers may agree separate deals that set terms for using their material, as some have done with Meta. Meta’s updated terms say automated tools can access and analyse pages linked to businesses using its services.
The change matters because it could affect how publishers’ work is used to develop AI, and whether they can set or negotiate terms for that use. Businesses that reject the new terms may lose access to Meta’s commercial products, while the terms are due to take effect on 30 October.
Both sides, in good faith
The strongest fair case each way — we don't pick a winner.
The case for
Meta could argue that analysing publicly accessible pages associated with businesses helps it improve AI tools and services used across its platforms, and that clearer, standardised terms make this access predictable. Businesses choose whether to use Meta’s commercial products, while separate licensing deals with some publishers show that negotiated arrangements can coexist with broader access.
The case against
Publishers could argue that using their reporting to train commercial AI without a separate licence or payment takes value from work they funded and may help create products that compete with them. Because businesses may have to accept the terms to keep using Meta’s commercial tools, they could see the choice to opt out as costly rather than freely given; the arrangement also raises concerns about consent and control over how their material is reused.