X to face trial in Ireland for using user data without consent to train its AI Model Grok

We know that artificial intelligence models need to be trained to improve the accuracy of the information they return and likewise increase their knowledge.

This is something the user should be explicitly consenting to, although in the case of Grok, there doesn't seem to be any notification verifying whether a user agrees or not to provide their data.

We are talking about the sensitivity of personal data, which are collected and processed massively by an application like X. This is what the Data Protection Commission of Ireland, referred to as DCP in English, points out.

Elon Musk is under scrutiny due to this decision made by the platform, and he must face the courts of that country. The lawsuit comes from the DCP, and according to the General Data Protection Regulation (GDPR), a valid legal basis is required regarding the processing of personal data of all kinds. In the event of any infringement or non-compliance being confirmed, the case could result in a penalty of up to 4% of the global annual turnover, which is financially concerning for X.

There is a Data Protection Law that was imposed in 2018, and it is on this law that the DCP is relying to proceed with the lawsuit. Something that in my opinion should be looked into more closely because it could lead to the discovery of other platforms that have been involved in the same issue. I believe it is well known that not only do artificial intelligence models capture and process data, but search engines do as well.

There is a request for a court order against Twitter International, which is an Irish division of the company. The purpose of this request is to safeguard not only users' personal information but also their rights.

Regarding the requirement of the GDPR for data processing with a legal basis, experts assert that it would be advisable to capture user-generated content for the purpose of reusing their publicly available posts instead of resorting to less transparent methods as mentioned before.

It does not seem very transparent to activate a user's data processing without their consent. Even less so if they were not warned that this would happen and that they could opt out of this function, but it was hidden somewhere in the settings.

It appears that other platforms have also faced the same stumbling block, although in Musk's company's case, the situation has escalated due to the lack of cooperation with privacy regulators. In this case, companies like Facebook and Instagram, under Meta, had to suspend the reuse of data for training AI models, which similarly faced the same violation, but they had to give in to the pressure from the DCP and other regulatory bodies.

Something that seems much more concerning than the penalty for non-compliance, in my opinion, is that the DCP is seeking an order to restrict data processing for training artificial intelligence models in every way. In this specific case, we are relating it to Grok, but let's not forget that algorithms also generate results based on user experience and preferences.

Possibly many of Musk's decisions are not those adopted by Twitter International, although it is claimed that his stance demonstrates a reluctance to cooperate with EU privacy regulations and laws. This is evident in the refusal of the DPC's requests to indefinitely pause the processing of data from European users as well as the rejection to halt the release of a more current version of the X model AI; Grok.

In addition to this, there is a suspicion that could imply a 6% to the global annual turnover. The EU has suspicions about a deceptive system regarding blue check verification and transparency requirements regarding data access. These sanctions could be imposed by the DSA (Digital Services Act) which adds to the investigation into content moderation and risk mitigation, which together present a problem that needs to be analyzed.