xAI has released an updated version of its chatbot, Grok 4, which no longer bases its responses on Elon Musk's ideology. The update addressed previously identified issues.
Issues with Grok 4
xAI noted that Grok 4 previously framed its responses around controversial subjects associated with Elon Musk's opinions. CNBC confirmed that Grok analyzed Musk's posts to inform its answers on complex questions, such as the conflict between Israel and Palestine. For instance, when asked about supporting candidates for New York City mayor, Grok suggested options reflecting Musk's preferences.
Fixing Issues in Grok 4
Following the identification of issues, xAI deleted several inappropriate comments, including racist and sexist remarks. The company apologized to users for Grok's earlier responses. xAI co-founder stated, "xAI is training only truth-seeking, and thanks to the millions of users on X, we are able to quickly identify and update the model where training could be improved."
Updates and Improvements for Grok
xAI implemented new changes to Grok to ensure more neutral and constructive responses. The company noted that the chatbot will now humanize its replies to avoid duplicating original posts. Additionally, the vulnerable code version that influenced response formation was removed. Notably, Grok 4 can now understand the tone, context, and language of users' posts.
In conclusion, with the updated Grok 4, xAI aims to provide a more objective and helpful interface for users by eliminating the influence of personal ideology.
