Skip to Main Content

Fermentation Sciences: Fermentation Sciences and ChatGPT

Getting Started with ChatGPT

ChatGPT is a large language model-based chatbot developed by OpenAI and launched on November 30, 2022, ChatGPT can assist with writing dialog, manuscripts, editing material, and creating new content. ChatGPT is one of several Generative AI platforms. Some others include: DALL-E-2, Bard, and Wordtune.

Check out Belk Library's Student Guide for ChatGPT

ChatGPT and Fermentation Sciences?

Could a tool like ChatGPT help me make some beer? Let's ask it and find out! Before we get started, it's important to note that ChatGPT is not thinking in the same sense you might consider a human might think. ChatGPT makes a lot of guesses as far as constructing language. It is pulling from a repository of information, and as the Student Tutorial above notes: ChatGPT was trained on human-written textual information available on the web, including conversations. For this reason, the response you receive often sounds life like. You are NOT talking to anyone real.

That being said, you, as the user, can get a result no matter how you interface with it. In this example, I have decided to talk to ChatGPT directly. Though, I could have also asked it like this: how can ChatGPT help me make a beer recipe?

Always feel free to ask ChatGPT directly what it can do! It's one of the fun features of this platform! Just remember that ChatGPT is limited in it's understanding of making beer (you will need to double check it's work, especially since ChatGPT is not shown to be proficient at research or intelligently reviewing data) More on this and other limitations in the Warning section below.

That's great--but, I need different fermentation information. Can ChatGPT help me understand microbial information and metabolic processes for different types of fermentations?

Again, let's ask it to get started:

Warning

While Generative AI, such as ChatGPT, is a great tool for revising or editing, remember that there is nothing secure about this outlet. Do not use sensitive information in your inquires (for example: consider omitting any personal information, or any other information where if others were to know, would result in unhappiness for you). 

Also, be aware of the ChatGPT self disclosed limitations:

  • ChatGPT sometimes writes plausible-sounding but incorrect or nonsensical answers. Fixing this issue is challenging, as: (1) during RL training, there’s currently no source of truth; (2) training the model to be more cautious causes it to decline questions that it can answer correctly; and (3) supervised training misleads the model because the ideal answer depends on what the model knows, rather than what the human demonstrator knows.
  • ChatGPT is sensitive to tweaks to the input phrasing or attempting the same prompt multiple times. For example, given one phrasing of a question, the model can claim to not know the answer, but given a slight rephrase, can answer correctly.
  • The model is often excessively verbose and overuses certain phrases, such as restating that it’s a language model trained by OpenAI. These issues arise from biases in the training data (trainers prefer longer answers that look more comprehensive) and well-known over-optimization issues.
  • Ideally, the model would ask clarifying questions when the user provided an ambiguous query. Instead, our current models usually guess what the user intended.
  • While we’ve made efforts to make the model refuse inappropriate requests, it will sometimes respond to harmful instructions or exhibit biased behavior. We’re using the Moderation API to warn or block certain types of unsafe content, but we expect it to have some false negatives and positives for now. We’re eager to collect user feedback to aid our ongoing work to improve this system. --Source, ChatGPT Blog

Science Librarian