houghts on Using ChatGPT o1
Translated by GPT
Thoughts on Using ChatGPT o1
I have been using the ChatGPT o1 model for a few days. I was amazed at how it specifically answers what the user wants and pinpoints the core compared to the previous model. For example, when asked, “Why did Heisenberg criticize Descartes’ philosophy?” the 4o model provided a lengthy explanation, whereas the o1 model summarized it concisely in one paragraph.
Both models output accurate information. However, there was a difference in their expression styles. Depending on how the same information is conveyed, it can either aid or confuse the user’s understanding. In this regard, o1 minimized misunderstandings by delivering information as logically as possible.
It has now become as smart as a human. It understands human intentions, provides clear answers, and doesn’t wander aimlessly like the AI we’ve seen before. In the near future, we might see an era where AI interprets the intentions of others, perhaps as a tool that aids communication through smart glasses.
Personally, it didn’t feel particularly refreshing. Perhaps the recent incredible performance of AI has made me indifferent. However, its logic and delivery have noticeably improved. If it continues to develop like this, I expect a future where AI makes decisions and humans only execute them.