Get your free copy of Editor's Digest
FT editor Roula Khalaf picks her favourite stories in this weekly newsletter.
Europe's data protection watchdog says it is “seeking clarification” over Elon Musk's X's decision to allow users' data to be automatically fed into his artificial intelligence startup xAI, stepping up new regulatory scrutiny of the social media platform.
X users discovered on Friday that they had “consented” to their posts on the site and their interactions with the Grok chatbot being used to “train and fine-tune” xAI's system.
This was done without explicit consent from users to share their data, and users cannot currently opt out from the mobile app as the setting can only be changed in the desktop version of X.
The Irish Data Protection Commission, which is responsible for ensuring internet companies comply with EU privacy law, said it had been in negotiations with Company X for “many months” about its plans to use user data to develop AI systems, and had also sent the company a letter of inquiry on Thursday about “transparency to users,” among other issues.
“We were surprised that they would initiate this,” a spokesman for the watchdog said, adding that the group had asked further questions about the matter on Friday “seeking clarification”.
Privacy experts have questioned whether X's actions violate the EU's General Data Protection Regulation, which requires companies that collect or use personal data to first get an individual's consent and disclose the reasons for doing so. If Irish regulators launch a GDPR investigation, X could face fines and penalties.
Last month, Meta suspended plans to train an AI with data from Facebook and Instagram platforms in Europe at the request of the Irish DPC over concerns about GDPR compliance, which Meta said represented a “setback for the race to innovate and develop AI in Europe.”
Musk bought X for $44 billion in 2022, but cuts to staff and moderation capacity and sudden product changes have led to increased regulatory scrutiny globally. The European Commission is already investigating the company for alleged transparency and moderation issues that violate the Digital Services Act.
“All X users can control whether their public posts are available to train our AI search assistant, Grok,” X's safety account wrote on Friday. “This option is in addition to the existing control over whether Grok-related interactions, typing, and results are available. This setting is available on the web platform and will roll out to mobile soon.” X did not respond to a request for comment.
The development comes as Musk seeks to distance himself from rivals such as OpenAI, Anthropic and Google, whose latest models he says are currently “orders of magnitude” better than xAI's. Access to X's vast repository of user data will be a key advantage for xAI, while Grok is available to X Premium subscribers.
Musk's network of companies is increasingly interconnected: Engineers and other senior staff are moving between his companies, and the serial entrepreneur is seeking approval from Tesla's board to invest up to $5 billion in xAI.
Some xAI investors point to synergies with Musk's other companies as a key advantage, but others worry about potential conflicts of interest.
