Grok is a large language model (LLM) created by Elon Musk’s company xAI, which works on a similar principle to the popular ChatGPT. This tool enables comprehensive text generation and other tasks related to natural language processing.
Elon Musk himself boasted that Grok is in many ways a better tool than ChataGPT, and at the same time “not devoid of a sense of humor” and “free from political correctness”
For a large language model to work well, it must be “fed” with gigantic data sets. This was the case with Grok itself, which was trained, among others, on content generated by X users. The problem is that this happened without their consent, or even informing them about this fact.
After disclosing this information, Elon Musk was met with criticism not only from users, but also from journalists, politicians and national privacy regulators. The Irish equivalent of our UODO, the Data Protection Commission (DPC), also spoke out on the matter.
. The Commission also announced that it had “already taken follow-up actions” and was waiting for a response from Elon Musk. However, it has not received this response.
X sued in Ireland. All because of Grok’s appetite
the Commission “has brought proceedings in the Irish High Court against Twitter International Unlimited Company over concerns about the way it processes the personal data of millions of European users of the ‘X’ platform”.
In an official statement, the DPC explained that its concerns center on the use of content generated by Portal X users residing in the European Union/European Economic Area to train artificial intelligence systems, including a tool known as Grok.
According to the Commission, such actions may violate the provisions of the GDPR. Let us recall that the general regulation on protection in force in the EU requires that any processing of personal data be based on a specific legal basis. Violation of the provisions of the GDPR may result in fines of up to 4% of the global annual turnover of a given company.
the DPC is seeking an injunction against Twitter (the Irish branch of the company has not changed its name to X) over threats to continue processing user data for the purpose of training AI models.
Grok is lying about the US election? In his world, Kamala Harris has no chance
This is not the end of the controversy related to Grok. A few days ago it turned out that the tool of Elon Musk, the candidate of the Democratic Party in the elections in the USA. Importantly, Musk himself officially supported her rival – Donald Trump – in the elections.
“We are calling on you to immediately implement changes to the xAI search assistant Grok to ensure voters have accurate information in this critical election year,” the secretaries of state of Minnesota, Pennsylvania, Washington, Michigan, and New Mexico wrote to Elon Musk.
Grok allegedly tried to convince his interlocutors that changing the Democratic candidate from Joe Biden to Kamala Harris was not possible in several US states because the deadline for voting on candidates had passed. In the meantime, it was still possible to change the candidate in each of the states, and the votes had not yet been held. Before the information was corrected, the chatbot repeated the lies for nine days.
“In an election year, it is critical that voters have accurate information about how to exercise their right to vote,” said Minnesota Secretary of State Steve Simon. “Voters should contact their state or local election officials to learn how, when and where they can vote,” added one of the signatories to Musk’s letter.
Source: Gazeta
Mabel is a talented author and journalist with a passion for all things technology. As an experienced writer for the 247 News Agency, she has established a reputation for her in-depth reporting and expert analysis on the latest developments in the tech industry.
