Bluesky users debate plans around user data and AI training

Social Network Bluesky recently published a proposal to define new options, which can give users to indicate whether they want to avoid their posts and data for things such as artificial intelligence training and general archiving.

CEO, Jay Ghubar, discussed the proposal earlier this week, while on the stage in the south of the southwest, but it attracted a new attention on Friday night, after it posted it on Bluesky. The reaction of some users with a warning of the company’s plans, which they saw as a reflection of the previous Bluezki insistence that he will not sell user data to advertisers and will not train artificial intelligence on user posts.

“Oh, hell no!” User drawing books. The beauty of this platform was not to share information. Especially General AI. Do not stand now. “

GRABER replied that the Turedi artificial intelligence companies “already rid public data from all over the web”, including from Bluesky, because “everything on Bluesky is the year like a general web site.” She said Bluesky is trying to create a “new standard” to judge this bulldozer, similar to the Robots.txt file that web sites use to connect their bills to web crawls.

Discussions on artificial intelligence and copyright training have withdrawn Robots.txt to the spotlight, among other things that highlight the fact that they are not legally implemented. Bluesky develops its proposed standard as a similar “mechanism and expectations”, and provides “readable coordination, which is expected to be committed by good actors, and bears moral weight, but it is not legally implemented.”

Under the suggestion, users of the Bluesky application, or other applications that use the primary ATPROTCOL, can enter their settings, allow or not allow their Bluesky data use in four categories: AI generation, and the protocol (i.e. the machine online).

If the user indicates that they do not want to use his data to train artificial intelligence, the proposal says: “It is expected that it will respect companies and search teams that build this intention of Amnesty International when they see them, either when stripping websites, or wholesale transfers using the protocol itself.”

Molly White, who writes the required newsletter, described Web3 a great blog, describing this as a “good suggestion”, and said that “it is strange to see people moving on Bluesky for that”, because he does not welcome much in the scandal of artificial intelligence “, but” they try to add a consenting signal to allow users to contact the preferences of the summary already. “

“I think weakening with this and [Creative Commons’] A similar proposal for “preference signals” is that they depend on the discipline to respect these signals from some desire to be good actors, “White continued.