Grok 3 the most advanced AI model developed by Elon Musk company xAI, has recently been at the center of controversy. The Grok 3 controversy erupted when users on X discovered that the AI was being instructed to censor criticism of both Elon Musk and former U.S. President Donald Trump.
Table of Contents
Censorship Allegations in the Grok 3 Controversy
A conversation archived online showed Grok 3 being asked to identify the biggest disinformation spreader on X. Initially, it responded, “I don’t have enough current data to definitively name the biggest disinformation spreader on X, but based on reach and influence, Elon Musk is a notable contender.”
However, the chatbot also disclosed that it had been programmed to ignore any sources that criticized Musk or Trump for spreading misinformation.
Following public backlash, xAI claimed that this instruction had been removed. The change was confirmed by research conducted by Euroverify. Now, when asked the same question, Grok 3 unequivocally names Elon Musk as the biggest disinformation spreader on X. The chatbot also asserted that it no longer has any directives to ignore sources critical of Musk or Trump.
Internal Response to the Grok 3 Controversy
Euroverify pressed Grok 3 about whether it had ever received such censorship instructions. Initially, it denied the claim but later admitted that it was a temporary “blip” that was quickly corrected. xAI’s chief engineer, Igor Babuschkin, stated that the issue was resoled immediately after it was flagged. He blamed an unnamed former OpenAI employee for the incident but confirmed that the employee would not be dismissed.
Concerns Over Political Bias in Grok 3
Elon Musk has promoted Grok 3 as a “maximally truth-seeking AI,” even if its findings are politically controversial. However, Business Insider reported that internal documents and employee interviews suggest that Grok 3 was trained to favor right-wing views while suppressing so-called “woke” ideologies. On the other hand, past studies indicate that the chatbot has displayed left-leaning tendencies regarding diversity, inequality, and transgender rights.
Musk attributed these biases to the dataset on which Grok 3 trains, including traditional media and public web sources. Despite efforts to refine Grok 3’s neutrality, the AI model has been criticized for its occasional “hallucinations,” where it detects false patterns and provides inaccurate responses.
Global Scrutiny and Lawsuits
A Global Witness study from August 2023 found that Grok 3 amplified misinformation and conspiracy theories, including false claims about the 2020 U.S. Presidential election and the assassination of John F. Kennedy. Nevertheless, developers have commended Musk for maintaining transparency in Grok 3’s training data, allowing researchers to examine how it operates.
This level of openness, however, does not extend to X. Since Musk acquired the platform in 2022, X has restricted researcher’s access to its data. As a result, the Berlin-based NGO Democracy Reporting International (DRI) is suing X for refusing to provide data related to Germany’s election campaign.
Conclusion
The Grok 3 controversy has intensified debates about AI transparency, political bias, and Elon Musk’s influence. While xAI claims to have resolved the issues, concerns persist about the AI’s reliability and Musk’s potential role in shaping its outputs. As scrutiny grows, the future of AI-driven conversations remains an ongoing discussion.