Accordingly, from July 2025, all domains using Cloudflare (US) services will by default deny access from AI data collectors, unless there is explicit consent from the site owner.
This new policy aims to prevent artificial intelligence (AI) companies from arbitrarily “scanning” text and image content from websites to train AI models without asking permission or paying the data owner.
Speaking about this pioneering decision, Mr. Matthew Prince, co-founder and CEO of Cloudflare emphasized: If the Internet is to survive in the age of AI, it needs to give control back to content creators, while still helping AI companies innovate and build a new economic model that works for everyone - creators, consumers, future AI founders and the future of the web itself.
For decades, content on the Internet has been created with the expectation that search engines will index it and direct users back to the original site, generating traffic and advertising revenue.
However, according to Cloudflare, this model is collapsing as many modern AI systems “suck” content such as text, articles, and images to generate answers without taking visitors to the original data source, causing creators to lose both revenue and motivation to create.
Cloudflare's policy not only makes it easy for websites to block AI crawlers with one click, but also forces AI companies to be transparent about how they use the data, such as model training, search, or inference, before asking for access.
Many major media and technology corporations around the world have supported Cloudflare's move. Mr. Roger Lynch, CEO of Condé Nast Group, said that this is an important step towards creating a fair exchange of value on the Internet that protects creators, supports quality journalism and holds AI companies accountable.
The entire ecosystem of creators, platforms, web users, and crawlers will be better off when data collection becomes more transparent and better controlled, said Steve Huffman, co-founder and CEO of Reddit.
With one of the largest networks in the world, Cloudflare now manages and protects traffic for 20% of the world’s websites. Since September 2024, the company has offered the option to block AI crawlers to more than 1 million customers. The next step in July 2025 is to make this option the default for all new domains, giving content owners control from the start.
Accordingly, AI companies will now have to get explicit permission from websites before collecting data. When signing up with Cloudflare, every new domain will be asked whether they want to allow AI crawlers, giving customers the choice from the start whether or not to allow AI crawlers access.
This change means that all new domains will be controlled by default and site owners no longer need to configure the opt-out themselves. Customers can easily check their settings and allow data collection at any time if they want their content to be freely accessible.
In addition, Cloudflare is also collaborating on developing a standard protocol that will help AI bots authenticate and websites identify these bots, creating conditions for the content ecosystem to become more transparent and accountable.
Source: https://nhandan.vn/ra-mat-dich-vu-dau-tien-tren-the-gioi-chan-ai-thu-thap-du-lieu-website-khi-chua-duoc-phep-post891533.html
Comment (0)