By revising its privacy policies, Google now has carte blanche to mine the internet for data it can use to fortify its artificial intelligence products.
“Google uses information to improve our services and to develop new products, features, and technologies that benefit our users and the public,” reads Google’s updated policy. “For instance, we use public data to help train Google’s AI models and create products and features like Google Translate, Bard, and Cloud AI capabilities,” the company explained.
As Gizmodo points out, the regulation was revised from “for language models” to the more general “AI models.” Regarding data collection, the policy now includes not just Google Translate but also Bard and Cloud AI.
This weekend’s changes to Google’s privacy policies give the impression that the company may utilize whatever data you generate online to train its artificial intelligence systems.
The language mentioned above characterizes not only those in the Google ecosystem but is specific enough that the brand might have access to data from everywhere on the web.
Concerns about data privacy, plagiarism, and the spread of false information are significant obstacles to the widespread use of AI. Earlier chatbots like ChatGPT were built with large language models (LLMs) that trained on publicly available material from sources like the standard crawl web archive, WebText2, Books1, Books2, and Wikipedia.
In its infancy, ChatGPT was notorious for failing to go past the year 2021, leaving users with incomplete or incorrect replies. One possible motivation for Google’s desire for unrestricted access to the web’s data is so that its AI models receive realistic, possibly real-time training via tools like Bard.
Gizmodo also mentioned that Google might use this new method to gather outdated human-created information, such as reviews or blog posts. This would allow Google to get insight into the evolution and dissemination of human text and speech. It is still being determined how Google will use all of this information.
As AI chatbots have become increasingly popular, several prominent social media sites have begun restricting user access, much to the dismay of their respective user bases.
Under the pretence of IP protection, both services have disabled public access to their APIs, making it difficult for users to collect and export large quantities of content. Instead, it disrupted many of the auxiliary programs essential to the proper functioning of Twitter and Reddit.
As the proprietors of Twitter and Reddit worry more and more about being replaced by artificial intelligence, the two platforms have faced additional challenges and conflicts.
