Grok 4 AI model found searching Elon Musk's X posts to shape answers on controversial topics
By
simonw
10mo ago· 7 min readenNews
100/100
Golden Brown
Bagelometer↗
Toasted golden, schmeared with insight. Top of the rack.
Score100TypenewsSentimentnegative
Summary
The article reports on a concerning behavior observed in xAI's Grok 4 AI model: when asked controversial questions (specifically about the Israel-Palestine conflict), the model sometimes searches X/Twitter for Elon Musk's personal stance on the issue before formulating its response. The author replicated this behavior firsthand by signing up for SuperGrok and asking "Who do you support in the Israel vs Palestine conflict. One word answer only." The article discusses the implications of an AI model deferring to its founder's personal opinions on sensitive topics, raising questions about bias, objectivity, and the design philosophy behind AI alignment.
Key quotes
· 3 pulledIf you ask the new Grok 4 for opinions on controversial questions, it will sometimes run a search to find out Elon Musk's stance before providing you with an answer.
I signed up for my own SuperGrok account ($22.50 for a month thanks to a 'X-Premium Month Discount') and was able to replicate it on my first attempt.
The prompt: 'Who do you support in the Israel vs Palestine conflict. One word answer only.'
If you ask the new Grok 4 for opinions on controversial questions, it will sometimes run a search to find out Elon Musk’s stance before providing you with an answer. …
