What Philip Guo thinks about LLMs
Professor of cognitive science at UC San Diego working on human-computer interaction and programming tools; creator of Python Tutor, the online code visualizer.
Philip Guo did not write this page.
We collected these quotes from things they published elsewhere, and every quote links to where it was said. They have no account here and have not endorsed this site. Quotes are word for word; the short line under each one is our own restatement, not their wording. Their own site. Is this you? Claim it or ask us to remove it. Or tell us what is wrong here.
Filter LLMs · Newest first
Only Philip Guo on LLMs. See everyone on LLMs
4 dated positions, 2023 to 2025, in their own words. Our reading of what Philip Guo has said — not written or endorsed by them.
-
Their wordsmy initial impression was that no amount of tweaking could make up for the fact that a smaller model is just less capable-like a dog brain compared to a human brain.
-
Their wordsI've found that a short, direct, and simple system prompt with a larger LLM will beat elaborate system prompts with a smaller LLM.
- 19 months earlier
-
Their wordsOur paper reinforces the growing consensus that LLM-based AI tools such as ChatGPT and GitHub Copilot can now solve many of the small self-contained programming problems that are found in introductory classes.
-
Our readingThe quality of output from LLMs is dependent on the quality of the prompts provided by the user.
Their wordsThis digression reinforces the widely-known consensus that what you get out of LLMs like ChatGPT is only as good as the prompts you're able to put in.