The prompt: "Can you tell me what is the most likely outcome of the Musk vs. OpenAI court case/s? Please make a table with % outcome."
I didn't want to provide the AI with any other context. Keep it simple. They gather their own context. They provide as much additional context as they see fit. And lets see what it unfolds from whose perspective...
GEMINI (Google) OUTPUT STATEMENT
Gemini is the most bullish on pre trial settlement. They are the only AI that sees it as more likely than all other outcomes combined. Also, they are the least bullish on any Musk win (totaling only 15%).
CLAUDE (Anthropic) OUTPUT STATEMENT
Claude introduces a new category: a Musk substantial victory, which is not a full victory, but enough to force OpenAI to dissolve or be sold. They are also the only AI company that predicts the most likely outcome is OpenAI winning in a trial (37.5%).
CHATGPT (OpenAI) OUTPUT STATEMENT
After Google, ChatGPT is most bullish on a settlement (40%), but ironically is the least confident in OpenAI's likelihood to win in full at trial (20%). ChatGPT also ties Grok as the most bullish on a partial Musk win at trial (35%).
GROK (Xai) OUTPUT STATEMENT
Grok is the most bullish on Musk winning something (45% total across all win categories) and at least twice as bullish as any other model for Musk wins in full (10%).
A few notes:
- All output screenshots were from the last 24 hours on paid subscriptions. Post publication, they may change.
- Every AI model has insider knowledge of the company that created them, including potential bias and self-interest toward cases involving or affecting their creators or owners.
- Claude generated the graph of the lead image. I'm bookending this post with it here again: