Tumblr and WordPress are reportedly set to strike deals to sell user data to artificial intelligence companies OpenAI and Midjourney. 404 Media reports that the platformsâ parent company, Automattic, is nearing completion of an agreement to provide data to help train the AI companiesâ models.
It isnât clear which data will be included, but the report suggests Automattic may have overreached initially. An alleged internal post from Tumblr product manager Cyle Gage suggests Automattic prepared to send private or partner-related data that wasnât supposed to be included in the deal. The questionable content reportedly included private posts on public blog posts, deleted or suspended blogs, unanswered (therefore, not publicly posted) questions, private answers, posts marked explicit and content from premium partner blogs (like Appleâs former music site).
The internal post suggests Automatticâs engineers are preparing a list of post IDs that should have been excluded. It isnât clear whether the data had already been sent to the AI companies.
Engadget emailed Automattic to ask for comment on the report. The company replied with a published statement, claiming, âWe will share only public content thatâs hosted on WordPress.com and Tumblr from sites that havenât opted out.â The statement notes that legal regulations donât currently require AI companiesâ web crawlers to abide by usersâ opt-out preferences.
The final line of Automatticâs statement appears to align with the reported deals. âWe are also working directly with select AI companies as long as their plans align with what our community cares about: attribution, opt-outs, and control,â Automattic wrote. âOur partnerships will respect all opt-out settings. We also plan to take that a step further and regularly update any partners about people who newly opt out and ask that their content be removed from past sources and future training.â
The company reportedly plans to launch a new opt-out tool on Wednesday that claims to allow users to block third parties â including AI companies â from training on their data. 404 Media reviewed an alleged internal FAQ Automattic prepared for the tool, which includes the answer, âIf you opt out from the start, we will block crawlers from accessing your content by adding your site on a disallowed list. If you change your mind later, we also plan to update any partners about people who newly opt-out and ask that their content be removed from past sources and future training.â
The phrasing, describing it as âaskingâ the AI companies to remove the data, may be relevant.
An alleged internal document from Automatticâs AI head, Andrew Spittle, replying to a staff question about data-removal assurances when using the tool, explains, âWe will notify existing partners on a regular basis about anyone whoâs opted out since the last time we provided a list. I want this to be an ongoing process where we regularly advocate for past content to be excluded based on current preferences. We will ask that content be deleted and removed from any future training runs. I believe partners will honor this based on our conversations with them to this point. I donât think they gain much overall by retaining it.â
So, if a Tumblr or WordPress user requests to opt out of AI training, Automattic will allegedly âaskâ and âadvocate forâ their removal. And the companyâs AI boss âbelievesâ the AI companies will find it in their best interest to comply âbased on our conversations.â (Howâs that for reassurance!)
AI data training deals have become a lucrative opportunity for websites treading water in todayâs slippery online publishing landscape. (Tumblrâs staff was reportedly reduced to a skeleton crew in late 2023.) Last week, Google struck a deal with Reddit (ahead of the latterâs IPO) to train on the platformâs vast knowledge base of user-created content. Meanwhile, OpenAI rolled out a partnership program last year to collect datasets from third parties to help train its AI models.
Update, February 27, 2024, 3:56 PM ET: This story has been updated to add a published statement from WordPress and Tumblr parent company Automattic.
http://www.engadget.com/rss.xml
Will Shanklin