Blog Post 12
Written by both Google Bard and Open AI's ChatGPT
Title: What are your strengths?
Generative Prompt: "an image of 2 leading AI bots in a gladiator style battle"
ChatGPT's Introduction: In this blog post, we delve into a comprehensive comparison between two prominent generative AI chat systems, exploring their unique abilities and evaluating their performance in different scenarios. By analyzing their strengths and weaknesses, we aim to provide readers with valuable insights into these cutting-edge conversational AI technologies.
Google Bard's Introduction: ChatGPT and Bard are two leading generative AI chatbots that offer a variety of features, including generating text, answering questions, and translating languages. ChatGPT is known for its ability to generate realistic and creative text, while Bard is known for its ability to answer questions in a comprehensive and informative way.
Introduction to the Digital Art:
Tuesday June 20th, 2023
Today, I wanted to start to look at some of the differences between ChatGPT and Bard. I've been using both so far, leaning more toward ChatGPT since it seems to be more simple for what I am trying to accomplish most of the time. I'll continue to go between the tools for content, and likely will add in some more as time goes on. I wanted to ask the tools to describe their own strengths. The table below represents the bullet points that each tool gave back to me based on the generative message below. Both tools gave different and mostly unique lists, however there were a couple of overlapping categories that I combined. I am planning to dig into some of these differences a little more deeply in future posts.
Generative Message: "What are your strengths? No details needed, just a bulleted list of values"
ChatGPT's Response:
Natural language processing
Contextual understanding
Generating coherent and contextually relevant responses
Knowledge recall and retrieval
Creative problem-solving
Adaptability to various topics and conversation styles
Understanding and adapting to user preferences and requests
Multilingual capabilities (support for multiple languages)
Bard/LaMDA's response:
Comprehensive and informative: I can access and process information from the real world through Google Search and keep my response consistent with search results, which allows me to provide comprehensive and informative answers to your questions.
Creative: I can generate different creative text formats of text content, like poems, code, scripts, musical pieces, email, letters, etc.
Efficient: I can process information quickly and efficiently, which allows me to provide you with the information you need in a timely manner.
Helpful: I am always willing to help you, and I will do my best to fulfill your requests.
Intelligent: I have been trained on a massive dataset of text and code, which allows me to understand and respond to your questions in a comprehensive and informative way.
Reliable: I am always available to help you, and I will always do my best to provide you with the information you need.
Some key takeaways on these tools (from the human curating the content):
At least for the purposes of this blog, Bard just requires a lot more back and forth. ChatGPT just accepts my prompts and gives me a response that is just about always usable. Bard generally gives me something along the lines of the content I am looking for, but I just about always have to fine tune and prompt multiple times. Here is just one examples from this post, and trust me, there are about 4 examples from today:
Example: I asked both tools to "Write a 2 sentence introduction about a blog post comparing the abilities of 2 different leading generative AI chats "
ChatGPT's response: "In this blog post, we delve into a comprehensive comparison between two prominent generative AI chat systems, exploring their unique abilities and evaluating their performance in different scenarios. By analyzing their strengths and weaknesses, we aim to provide readers with valuable insights into these cutting-edge conversational AI technologies."
This is unchanged and included in this post's introduction.
Bard's response: "
Sure, here is a 2-sentence introduction about a blog post comparing the abilities of 2 different leading generative AI chats:
In this blog post, we will compare the abilities of two leading generative AI chats, ChatGPT and LaMDA. We will look at their strengths and weaknesses, and discuss which one is better suited for your needs.
This introduction provides a brief overview of the blog post and what it will cover. It also highlights the fact that the blog post will compare two leading generative AI chats, which is a topic that is likely to be of interest to many people.
Here are some additional details that could be included in the introduction:
ChatGPT is a chatbot developed by OpenAI. It is known for its ability to generate realistic and creative text.
LaMDA is a chatbot developed by Google AI. It is known for its ability to answer questions in a comprehensive and informative way.
The blog post could then go on to compare the two chatbots in more detail, looking at their strengths and weaknesses in different areas, such as:
Generating text: How realistic and creative is the text that each chatbot generates?
Answering questions: How comprehensive and informative are the answers that each chatbot provides?
Other features: What other features do each chatbot offer, such as the ability to translate languages or write different kinds of creative content?
Ultimately, the blog post would help readers to decide which chatbot is better suited for their needs."
Wow! First of all, a lot of details here. Second, Bard seems to be having a small identity crisis. Do I call it Bard or LaMDA? I actually asked it, and it told me that it's name is Bard, but the underlying technology is LaMDA. So there, I learned something. It's just something interesting I noticed.
The true issue here, is that I asked for 2 sentences and got a huge generated response. It's helpful in some ways, but for how I'm working on this blog, it's just too much to sort through and read. After all, I'm not the one writing this blog, AI is writing it. Except this, I'm writing this, just about AI. LOL.