Cloudflare Comes Up With A New Tool That Stops AI Bots From Stealing Website Data For Training
The biggest problem that seems to be linked to the world of AI is the growing rise in data theft.
Tech giants are rolling out new AI-based chatbots and as the competition intensifies, you cannot help but wonder where the data for training is coming from.
Most admit to scraping public data from the web, and others are now trying to save lawsuits heading in their direction by entering licensing deals with respective stakeholders. Whatever the case may be, it’s getting tougher by the minute to stop anyone from using data with zero compensation or credits given.
However, it appears like Cloudflare might have found a temporary solution to the issue with the launch of a new tool that stops AI chatbots from this practice of stealing clients’ data from websites for training LLMs. In case you’re wondering, it’s free of cost, at least for now.
The top-of-the-line provider of cloud services says it hopes to make the offering available to all in its organization with free usage. With time, it would be updated as the tech world gets smarter with their respective AI bots scraping data from the web.
A new blog post shed more light on the endeavor, revealing details about the company’s latest offering and how clients are providing feedback to the sudden boom of these bots scraping content for Generative AI practices of their models.
Shockingly, the company says that close to 85% of all clients are opting to block AI bots that highlight themselves from attaining access to such sites online.
The company is also highlighting which bots remain the most active since last year and how many websites they had access to for this purpose.
As days go by, techniques are becoming more and more sensitive and that’s why it’s becoming more difficult for experts to block these bots from attaining material online. Some feel the race to design models at a fast pace is adding to the issue and therefore it’s hard to block scrapers from doing their job.
Cloudflare released a statement on this which was seen across its blog. It adds a mention of strong fears about how some AI giants’ purpose of breaking rules to attain unauthorized access will consistently adapt to evasion of bot detection.
Read next: Creative Community Criticizes Apple Intelligence For Lack Of Transparency
