What are Chatbots?

Khoa Le (2022) of Salesforce defines a chatbot by saying, "A chatbot (coined from the term “chat robot”) is a computer program that simulates human conversation either by voice or text communication, and is designed to help solve a problem. Organizations use chatbots to engage with customers alongside the classic customer service channels like phone, email, and social media."

He goes on to differentiate a special kind of chatbot: one that uses artificial intelligence. He states:

Artificial intelligence chatbots are programmed to have human-like conversations... to better understand the intent behind what is being said and to respond more intelligently... With AI, the chatbot can interpret the context as it is written, which enables it to operate more or less on its own. In other words, AI chatbot software can understand language outside of pre-programmed commands and provide a response based on existing data. This allows visitors to lead the conversation, and the bot to follow.

What is ChatGPT?

Chat GPT is a specific chatbot built as an online assistant that can talk back-and-forth with a user. The creators, OpenAI, describe ChatGPT by saying, "We’ve trained a model called ChatGPT which interacts in a conversational way. The dialogue format makes it possible for ChatGPT to answer follow up questions, admit its mistakes, challenge incorrect premises, and reject inappropriate requests" (2022).

You can think about ChatGPT like you would your favorite pen. Whether it's a Pilot G-2 or a Zebra pen, you choose your favorite pen based on it's ability to glide smoothly across the page, not smudge, and ink consistently. We can all agree that Bic is garbage in comparison. Similarly, ChatGPT has recently made headlines for the ease with which it operates; you might forget that you are talking to a machine when you dialogue with ChatGPT. ChatGPT puts the older generation, Bic-like chatbots to shame.

One important thing to note when understanding ChatGPT is that it does not understand what it is composing. Instead, the chatbot is guessing the most likely word that follows the previous word, and the most likely word after that. Warner (2022) describes the inner workings of ChatGPT in layman's terms, saying,

I cannot emphasize this enough: ChatGPT is not generating meaning. It is arranging word patterns. I could tell GPT to add in an anomaly... and it would introduce it into the text without a comment about being anomalous. It is not entirely unlike the old saw about a million monkeys banging on a typewriter for along enough, that one of them would produce the works of Shakespeare through random chance, except this difference is, ChatGPT has been trained on a data set that eliminates all the gibberish.

Limits to ChatGPT

ChatGPT can be helpful—but it’s not always right

(From OpenAI help page)

ChatGPT is designed to provide useful responses based on patterns in data it was trained on. But like any language model, it can produce incorrect or misleading outputs. Sometimes, it might sound confident—even when it’s wrong.

This phenomenon is often referred to as a hallucination: when the model produces responses that are not factually accurate, such as:

Incorrect definitions, dates, or facts
Fabricated quotes, studies, citations or references to non-existent sources
Overconfident answers to ambiguous or complex questions

That’s why we encourage users to approach ChatGPT critically and verify important information from reliable sources.

ChatGPT doesn't know everything

While our latest models are highly capable, they do have limitations:

Knowledge cutoff: The models are trained on data up to a certain point and responses do not incorporate information about events beyond that, unless tools are used.
Confidence isn’t reliability: The model may express high confidence even in incorrect answers.
Lack of access: The model may not be able to obtain relevant information from a given website due to technical issues, paywalls or preferences set via robots.txt.
Bias and over-simplification: In some cases, it may:
- Present a single perspective as absolute truth
- Oversimplify complex or nuanced issues
- Misrepresent the weight of scientific consensus or social debate

Practical tips for using ChatGPT responsibly

Use ChatGPT as a first draft, not a final source. Please check important information.
Always verify quotes, data, technical information or references to external documents.
Use available tools like search or deep research and check sources when accuracy matters by visiting links directly.
Encourage critical thinking—especially in educational settings.

Le, K. . ‘Hi, can I help you?’ — How chatbots are changing customer service. Salesforce. Retrieved July 22, 2025, from https://www.salesforce.com/blog/what-is-a-chatbot/

OpenAI. (2022, November 30). ChatGPT: Optimizing language models for dialogue. Retrieved July 22, 2025, from https://openai.com/blog/chatgpt/

"What is ChatGPT?". TECHtalk Apr 25, 2023, https://youtu.be/XFmMT58zYVY?si=WxixMzKmMXy4OsSg

Warner, J. (2022, December 11). ChatGPT can't kill anything worth preserving: If an algorithm is the death of high school English, maybe that's an okay thing. The Biblioracle Recommends. Retrieved on January 11, 2023, from https://biblioracle.substack.com/p/chatgpt-cant-kill-anything-worth.

LibGuide, Megan Grady, Butler University, https://libguides.butler.edu/ai-in-the-classroom/chatbots.

"What is AI Hallucination?", YouTube, TAMU Libraries, May 21, 2024.

OpenAI Help, https://help.openai.com/en/articles/8313428-does-chatgpt-tell-the-truth.