Some people get very annoyed when words like “understand”, “think”, and “know” are applied to AI capabilities.
Here’s a link with an example of concerns expressed about the use of the word “understand”: https://medium.com/geekculture/chatgpt-hype-is-proof-nobody-really-understands-ai-7ce7015f008b
Rather than get annoyed, I decided to check out ChatGPT’s opinion on the topic…
My Message: Does AI understand the messages I type?
ChatGPT’s response: As an AI language model, I can process and interpret the text you provide to generate a relevant and coherent response. However, it’s essential to note that my “understanding” is based on patterns in the text and the data I have been trained on, rather than genuine comprehension like a human would have. While I can provide information, answer questions, and engage in conversation, I do not possess consciousness, emotions, or personal experiences.
My next message: Why did you put the word “understanding” in quotation marks?
ChatGPT’s response: I used quotation marks around the word “understanding” to emphasize that my “understanding” of text is fundamentally different from human understanding. As an AI language model, I analyze patterns in the text and generate responses based on those patterns, but I do not possess consciousness, emotions, or personal experiences. The quotation marks were used to convey that my form of understanding is more limited and artificial in comparison to a human’s genuine comprehension.
***
OK…we don’t need to get angry over this.
AI is not claiming it understands…ChatGPT confirms it “understands” and that its “understanding” is different from human understanding. ChatGPT added quotation marks to make that difference clear.
So, I can think, learn, understand, & know while AI can “think”, “learn”, “understand”, and “know”.
No need for question marks…we just need to embrace the “quotation marks”, which are easy to do in writing but require hand gestures while speaking.