Ask HN: Generative AI and “fuzzy programming”
Fuzzy Logic is a logic based on an imprecise value for true and false. It seems to me that from a perspective of imprecise concepts and conditions, LLMs allow the same feature for programming: Instead of stating the exact method of measurement and the exact threshold, we can now describe a condition intuitively in plain language and rely on an LLM such as ChatGPT to determine if the condition applies to the current context. We can thus do "fuzzy programming" with algorithms based on imprecise categories, conditions, and entities. Are there any technical papers or blogs that folks can recommend that talk about the unlocked capabilities of fuzzy programming? I am especially interested in design patterns and any discussions on features that are greatly improved by imprecise programming.
0 comments
—