Skip to Main Content

Using AI Language Models

A brief overview of how to use AI Language Models

GPT

What is ChatGPT?

ChatGPT is designed to understand natural language and respond in a way that feels natural and conversational. It is built on a complex neural network that’s been trained on an enormous amount of information, helping it recognize patterns and generate thoughtful replies.

When someone types in a question or statement, ChatGPT works through a process called natural language processing—or NLP, for short. It essentially breaks the text into smaller parts, figures out how those pieces relate to one another, and then crafts a response that sounds more like how a person might talk.

Its training data comes from various sources, including books, articles, websites, and other texts representing many different writing styles. This allows it to generate text on various subjects, from scientific and technical topics to social and cultural issues.

One of the advantages of its architecture is that It can continue learning and improving over time. As more data becomes available, It can be retrained on this data to improve its language processing and generation capabilities. It can also learn from user feedback and adjust its responses to better meet their needs and expectations as more users use it over time.

What Does GPT Mean?

GPT stands for "Generative Pre-trained Transformer." It is a type of neural network architecture used in natural language processing tasks, such as text generation, translation, and question-answering. The models are designed to generate text similar to everyday human language, using complex algorithms (mathematical procedures) applied to large amounts of data. The GPT models are part of a family of neural network bots that have become increasingly popular in many applications over the past few years, and no doubt, these models will be added to more and more areas of everyday life. 

Applications and Impact:

ChatGPT 3 has already shown potential in numerous industries and applications, including:

  • ChatGPT's natural language processing abilities can improve chatbots and virtual assistants that can respond to human language in a more natural way.
  • The models can also generate content, including articles, blogs, and even attempt poetry (the results are debatable as to its ability to create poetry), which makes it useful for some forms of content creation tasks. This model can offer insights and ideas for structuring and organizing written work, generate prompts and provide feedback to help them improve their vocabulary and grammar, and create study aids, such as flashcards or quizzes.
  • ChatGPT's ability to translate between different languages has the potential to make language barriers less intimidating, possibly providing new opportunities for global communication and collaboration.
  • ChatGPTs can create personalized learning experiences for individual needs and interests. It can suggest activities based on preferences and learning styles; this might help users to stay engaged and motivated.
  • Regardless of whether the output was produced using a free or paid plan, you own the output you make with ChatGPT, including the right to reprint, sell, and merchandise, subject to their Content Policy and Terms.

ChatGPTs can potentially change how we interact with machines and understand languages. Its features and capabilities make it an interesting tool for various industries and everyday functions, and we are just beginning to see what it can do.

Possible Downsides:

There are several potential downsides to consider:

  • ChatGPTs may produce discriminating or biased (in favor or against something) outcomes if the bots are trained on skewed data. Models can pick up and reflect biases in their replies. If the human trainers are biased (have a preference), ChatGTPs will serve as a vehicle for spreading those beliefs, and users who do not undertake their own research may come to similar conclusions.
  • The inner workings of ChatGPT are not fully transparent, which makes it highly problematic to know exactly how the model arrives at its output. This lack of transparency can make it difficult to trust or verify the output, particularly in things such as sensitive healthcare or legal decision-making. Always keep this in mind when working with Chatbots. They do not allow you to know who is giving the bot its information. 
  •  The ease of generating content using ChatGPT may lead to an overreliance on bots, potentially impacting human employment and education. This could have future implications for job security, the quality of content produced, and how we learn. 
  • If students use ChatGPT to generate essays or other academic assignments without citing them, this could lead to academic dishonesty and undermine the learning process.

  • Always check with your instructor to find out if ChatGPT can be used as a scholarly resource.   
  • Students relying heavily on ChatGPT to generate content may miss out on the critical thinking and analysis skills essential for learning. By relying on a machine to do their work, they may not develop the ability to evaluate information critically, form their ideas and arguments, or build complex concepts. The world needs "individual voices" to be a better place! What would the world look like if individuals relied on AI to express themselves?! 
  • It is important to emphasize that while ChatGPT can be a helpful tool for users, it should not be relied upon as a substitute for human instruction or interaction. It may be able to provide information and generate responses, but it cannot provide the same level of feedback and guidance that a human instructor can. 
  • ChatGPT may not always be able to understand the question or response, which could lead to misunderstandings or inaccurate information being provided. It can provide wrong answers! Always check the information that you are being given when using a bot. The user is responsible for checking all of the information it provides. 
Check Out These Great Links! 

ChatGPT sets record for fastest-growing user base - analyst note

Source: HU, Krystal (2023-02-02) “ChatGPT sets record for fastest-growing user base - analyst note”. Reuters. 

Where is ChatGPT taking us? And do we want to follow? | Hub (jhu.edu)

Source: Hub (2023-02-09). “Where is ChatGPT taking us? And do we want to follow?” Interview with Daniel KHASHABI. Johns Hopkins University.

Reinventing search with a new AI-powered Microsoft Bing and Edge, your copilot for the web - The Official Microsoft Blog

Source: MEHDI, Yusuf (2023-02-07). “Reinventing search with a new AI-powered Microsoft Bing and Edge, your copilot for the web”. 

Why We're Obsessed With the Mind-Blowing ChatGPT AI Chatbot

Source: Sharkland, S. (2022, December 22). Why everyone's obsessed with ChatGPT, a mind-blowing AI chatbot. CNET. Retrieved January 5, 2023.

OpenAI

GPT 4 - Technical Report by OpenAI

ChatGPT | OpenAI Help Center by OpenAI

How To Use ChatGPT For Keyword Research

Source: Smullen, D. (2023). How To Use ChatGPT For Keyword Research. Search Engine Journal. https://www.searchenginejournal.com/chatgpt-for-keyword-research/483848/#close

On CRM: Forget ChatGPT, This Is How AutoGPT Will Soon Disrupt Your CRM System

Source: Marks, G. (2023, April 19). On CRM: Forget ChatGPT, This Is How AutoGPT Will Soon Disrupt Your CRM System. Forbes. https://www.forbes.com/sites/quickerbettertech/2023/04/19/on-crm-forget-chatgpt-this-is-how-autogpt-will-soon-disrupt-your-crm-system/?sh=7e0e2b935b59

Here comes Bard, Google’s version of ChatGPT

Source: Ghaffary, S. (2023, February 6). Google unveils Bard, its AI chatbot to rival ChatGPT. Vox. https://www.vox.com/recode/2023/2/6/23588308/google-bard-chatbot-chatgpt-ai-testing-public

Is the future about one all-knowing AI or many? The new app Poe gets you ready to chat with them all.

Source: Benton, J. (n.d.). Is the future about one all-knowing AI or many? The new app Poe gets you ready to chat with them all. Nieman Lab. https://www.niemanlab.org/2023/02/is-the-future-about-one-all-knowing-ai-or-many-the-new-app-poe-gets-you-ready-to-chat-with-them-all/

AI-Powered Text to Voice Generator

Source: Play.ht. (n.d.). AI Voice Generator & Realistic Text to Speech Online. https://play.ht/?via=nick

DALL·E 2 - AI system that generates images

Source: DALL·E 2. (n.d.). https://openai.com/product/dall-e-2

Microsoft and OpenAI Extend Partnership

Source: Blogs, M. C. (2023, February 7). Microsoft and OpenAI extend partnership - The Official Microsoft Blog. The Official Microsoft Blog. https://blogs.microsoft.com/blog/2023/01/23/microsoftandopenaiextendpartnership/

  AI Index Report -Standford University

Source: Artificial Intelligence Index - Stanford Institute for Human-Centered Artificial Intelligence

GPT-5