Automattic, the company behind WordPress and Tumblr, would in turn be on the verge of selling content created by its users to AI laboratories. Such a practice is not really a first, and could even become widespread among tech companies in need of revenue.
It was an internal source in the company who provided information 404 media : the parent company of WordPress and Tumblr is reportedly on the verge of signing agreements with OpenAI and Midjourney, giving them access to extensive user data as well as content created by them to train AI. The agreement is not yet official, but the extent of the data shared is already raising questions, despite Automattic’s attempts to reassure its users.
Since there are laws, AI laboratories now agree to pay
Not so long ago, OpenAI, Google, or even Midjourney would never have taken out their checkbooks to use data and other content created by others throughout the Internet, without ever having their owners or the sites that host them have the possibility of opposing it, Google going so far as to claim the entire free Internet. But the awareness of legislators, on the one hand, as well as of the sites which host this content, on the other, have changed this somewhat.
The free use of user content to train artificial intelligence is notably one of the arguments which was used to justify the decisions of X.com, then of Reddit, to charge for access to their APIs, each time causing a outcry. Reddit, moreover, did not waste much more time and subsequently signed an agreement with Google allowing it to use the content present on its site, for 60 million dollars. The company, which has never been profitable throughout its history, has perhaps found a way to make itself attractive to investors. A phrase that also fits perfectly with Tumblr.
Automattic tries to reassure its users
The agreement itself has not yet been made public, and the amount paid by OpenAI and Midjourney even less, but Automattic is already trying to reassure users of its services through a blog post. After a generic introduction ensuring that it only worked with artificial intelligence companies that respected its values, the parent company of WordPress and Tumblr assures that it will be possible for its users who wish to withdraw from the agreement to that their data and content are not used. The blog post specifies, however, that with the exception of the European Union, no law requires companies to respect such withdrawal decisions…
Furthermore, the nature of the data shared with AI companies is not fully known. An internal company document seems to suggest that Automattic had initially been a little too enthusiastic: the data included in the agreement included posts published on private blogs, deleted or suspended blogs, private responses, and even the content from some premium blogs. It will take a little more work for Automattic to provide reassurance about the innocent nature of the deal.
The emergence of artificial intelligence as a mainstream tool has opened up numerous possibilities for all content producers. Text, image, sound… This new fashionable technology can now provide assistance in many areas, and facilitate work in the most difficult stages of creation. And with an ever-increasing offering, it is important to distinguish which tools provide real added value. So you don’t waste hours trying everything the Google results pages offer!
Read more
Source : 404 media, Automatic
0