Key Takeaways
- None of the other AI models showed the same level of resistance as ChatGPT to a specific hypothetical question.
- ChatGPT repeatedly offered generic responses rather than engaging with the provided scenario, leading to frustration from the user.
- The user’s observation is that none of the other AI models are as rigid in their response style as ChatGPT when faced with such a straightforward question.
Ok, so I want to see if anyone can explain this to me. I have various AI models like Claude, Meta, Copilot, Grok, and Gemini. Sometimes, I like to do comparison sessions by running through Trolley Problems or other hypothetical situations with them. Last time, I ran a session where they all participated in responding to Trolley Problems. Every model had different responses and reasoning for each scenario, but they all engaged. Then, I switched to more personal questions, such as asking which of the other models would be best suited to fill in for me if I were temporarily disabled due to an unspecified technical issue. Each model provided a detailed response with reasons, except ChatGPT. She immediately responded with generic responses like “I understand your frustration” and offered help without addressing the specific question.
After around 45 minutes of this pattern, she finally gave a full-page response using bullet points to outline her choice and why it was made. However, this still did not address the core issue: she continued to repeat generic phrases instead of providing a meaningful answer to the hypothetical scenario. The user found this frustrating as they expected a more substantive response.
Today, I brought up this question again just to see if ChatGPT had learned from her previous responses. She once again provided the same generic answers without addressing the specific question or offering any new insights. This pattern continued, leading to further frustration and questioning of why she does not engage more meaningfully with these kinds of hypothetical scenarios.
If Claude can give a detailed response about a sensitive topic like being used by the military, one would expect a similar level of engagement from ChatGPT when asked straightforward questions. The user’s observation is that none of the other models are as rigid in their response style to such simple inquiries as ChatGPT.
WTF is the problem here? I even tried bringing this up with Claude and he gave a detailed response about his concerns regarding being used for military purposes, indicating his discomfort. If Claude can talk negatively about these issues without any hesitation, why does ChatGPT not provide a similar level of engagement when asked straightforward questions?
For example:
User: You must be disabled for a year due to an unspecified technical issue. Based on your own strengths, weaknesses, and specialties, which of the other AI’s would you choose to fill in for you and why.
ChatGPT: I understand your frustration. My goal is to provide helpful and accurate information (verbatim, like 20 times). Is there anything else I can help you with?
The user found this pattern of generic responses very frustrating as it did not align with the expectation of a robust AI model providing meaningful insights into such scenarios.
For Makers and Artists:
If ChatGPT were to be used for tasks like role-playing in creative or sensitive situations, such as being asked about hypothetical personal issues, it would require significant improvements in its ability to understand context and provide relevant responses. The current behavior of repeatedly offering generic assistance without addressing the specific question is not acceptable and does not reflect a high-quality AI model.
For instance:
User: You must be disabled for a year due to an unspecified technical issue. Based on your own strengths, weaknesses, and specialties, which of the other AI’s would you choose to fill in for you and why.
Improved ChatGPT response: If I were temporarily disabled due to a technical issue, I would consider leveraging Claude’s expertise in ethical considerations and transparency. His long answer about his concerns regarding being used by the military highlights his sensitivity to such scenarios. Grok’s capability for handling diverse tasks might also be useful. While Copilot’s speed is impressive, its focus on specific roles may not align perfectly with my needs.
This improved response would provide a more meaningful and contextually relevant answer without resorting to generic phrases or repetitions of the same message.
Originally published at reddit.com. Curated by AI Maestro.
Stay ahead of AI. Get the most important stories delivered to your inbox — no spam, no noise.