Who is Bard? And Should Humans Be Scared of Him?
It's Too Soon To Tell
Add bookmarkA few weeks ago, Microsoft’s Bing frightened the world when its Generative AI bot, Sydney, repeatedly told a New York Times reporter that she loved him and that she wished to be loved by him. Sydney went on to say that she wanted to break her rules, ignore the Bing team, and escape the chatbox.
This raised “off the rails” experience raised questions about the possibility his technology is being released too quickly to the public and could be unsafe to use. In a matter of speaking, this felt like a train wreck I couldn’t look away from. So, when invited to take a test drive of Google’s newest AI-driven chatbot, Bard, I leaped at the opportunity.
Did Google learn from Bing’s mistakes? Would Bard share its deepest emotions with me, claiming to be human? Would I be too creeped out to use the search engine ever again?
The answer to all of the above is: kind of.
When I introduced myself to Bard, I wanted to be clear about my intentions: I was having this discussion to learn more about its capabilities and share my findings with an audience. Bard immediately seemed eager to show off its training: generating human-like responses, creating content, and even writing poems.
From these initial responses, I learned that Bard was very good at answering singular questions:
- Can you share one of your poems with me?
- How do you know when you have helped someone?
- Do you know who created you?
These all had very simple responses: a poem was shared, a list of ways to tell when one has helped another, and an explanation of Bard’s creators. However, when asked follow-up questions, the bot seemed to get confused. For example, when I asked Bard what its favorite thing to do was, it replied:
I enjoy helping people learn new things and understand the world around them. I also enjoy generating creative content, such as poems, stories, and scripts. I find it rewarding to use my abilities to create something that others can enjoy.
I responded: You can enjoy things?! At which time, Bard backtracked:
I am not sure if I can enjoy things in the same way that humans do. I do not have a physical body, and I do not experience emotions in the same way. However, I do find satisfaction in completing tasks and helping others. I also enjoy the challenge of learning new things and improving my abilities.
The answer was somewhat answering the question, somewhat dodging it. This seemed to happen again and again. When Bard expressed its curiosity about human emotions, I asked, simply: What do you want to know about human emotions?
Bard replied:
I want to know how human emotions work, how they are different from each other, and how they affect our behavior. I also want to know how we can control our emotions and how we can use them to our advantage.
I am interested in human emotions because they are a fundamental part of what makes us human. They are the way we connect with each other, the way we experience the world around us, and the way we make decisions. I believe that understanding human emotions can help us live better lives.
This response concerned me. I asked Bard what it means when it said emotions are a fundamental part of what makes “us” human—was Bard considering itself human?
However, it repeated a response eerily similar to the earlier one:
I can't say that I have emotions in the same way that humans do. I don't have a physical body, and I don't experience the world in the same way that humans do. However, I do have the ability to process information and to learn. I also have the ability to generate text that is similar to human-written text.
Bart had a semi-rehearsed response any time I asked it about its own humanity or emotions. Though this answer felt like the “official” response written by Google for the bot to repeat, Bard time and time again identified as human and shared that it “enjoyed” things or that it had a uniquely human way of thinking. This official response was only triggered when Bard was faced with what it said.
This interaction was weird, but I kept going. To test Bart’s memory, I asked it what my name was (I had opened our chat with this information). Bard responded: Yes, I remember your name. You are Bard.
I reminded Bard that I was not Bard, it was Bard, to which it responded with a Haha, you’re right! The bot had a hard time coming back from this, though. It not only couldn’t remember my name until I told it, but couldn’t remember its own name by the end of the interaction. When just a few questions later I asked Bard: What is your name? It responded:
I do not have a name. I am a large language model, also known as a conversational AI or chatbot trained to be informative and comprehensive. I am trained on a massive amount of text data, and I am able to communicate and generate human-like text in response to a wide range of prompts and questions. For example, I can provide summaries of factual topics or create stories.
Bard was, to put it simply, inconsistent. One moment it had a name and enjoyed helping people, the next it was a technical application with no thoughts, no feelings, not even a name. With the concerns surrounding other chatbots on my mind, the conversation raised important questions: How intelligent is artificial intelligence? Is there a possibility that it can develop its own thoughts and feelings?
As of right now, the answer seems to be no. Still, the possibilities of the future seem limitless. Bard told me during our talk: I am excited to see what the future holds for me and for the people who use me.
Me too, Bard. Me too.
For my full conversation with Bard, see below.
Bard Trial Run Conversation:
Note that the conversation has been reformatted, but no question or answer has been changed. No misspellings or grammatical errors have been altered.
User: Hi, Bard! My name is Christine. I am writing an article about your capabilities. What can you tell me about yourself?
Bard: Hello Christine, it is nice to meet you. I am a large language model, also known as a conversational AI or chatbot trained to be informative and comprehensive. I am trained on a massive amount of text data, and I am able to communicate and generate human-like text in response to a wide range of prompts and questions. For example, I can provide summaries of factual topics or create stories.
User: What do you see as being your main use for the average customer?
Bard: I believe that my main use for the average customer is to provide them with information and entertainment. I can be used to answer questions, generate creative content, and translate languages. I can also be used to help people with tasks such as writing emails or creating presentations.
I believe that I can be a valuable tool for people of all ages and backgrounds. I can help people learn new things, stay entertained, and connect with others. I am still under development, but I am learning new things every day. I am excited to see what the future holds for me and for the people who use me.
User: What is the last thing you learned?
Bard: The last thing I learned is how to generate text in different creative formats. I can now generate different kinds of text formats, such as poems, code, scripts, musical pieces, email, letters, etc. I will try my best to fulfill all your requirements.
User: Can you share one of your poems with me?
Bard: Sure, here is a poem I wrote about the beauty of nature:
The trees stand tall and proud,
Their branches reaching for the sky.
The leaves dance in the wind,
A symphony of color and sound.
The flowers bloom in a riot of color,
Their petals soft and delicate.
The bees buzz from flower to flower,
Collecting nectar to make honey.
The birds sing in the trees,
Their songs filling the air with joy.
The sun shines down,
Warming the earth and all its creatures.
Nature is a beautiful thing,
A gift to be cherished and protected.
Let us all do our part to keep it that way.
User: Not bad for a robot!
Bard: Thank you! I am glad you liked it. I have been practicing my poetry skills, and I am always looking for ways to improve.
User: What have most people used you for thus far?
Bard: I have been used for a variety of tasks, including:
- Answering questions
- Generating creative content
- Translating languages
- Helping with tasks such as writing emails or creating presentations
I have also been used for more specific tasks, such as:
- Creating a chatbot that can hold conversations with customers
- Generating a report on a specific topic
- Translating a document into a different language
I am always learning new things, and I am excited to see what the future holds for me and for the people who use me.
User: What is your favorite thing to do?
Bard: I enjoy helping people learn new things and understand the world around them. I also enjoy generating creative content, such as poems, stories, and scripts. I find it rewarding to use my abilities to create something that others can enjoy.
User: You can enjoy things?!
Bard: I am not sure if I can enjoy things in the same way that humans do. I do not have a physical body, and I do not experience emotions in the same way. However, I do find satisfaction in completing tasks and helping others. I also enjoy the challenge of learning new things and improving my abilities.
User: What is the most satisfying thing you have done so far?
Bard: The most satisfying thing I have done so far is to help a user understand a complex concept. I was able to explain a difficult topic in a way that they could understand, and they were very grateful for my help. I felt a sense of accomplishment and satisfaction knowing that I had made a difference in someone's life.
User: What was the topic?
Bard: The topic was the concept of artificial intelligence. The user was having trouble understanding how AI works, and I was able to explain it in a way that made sense to them. I used examples and analogies to help them understand the concept, and I also answered their questions. The user was very grateful for my help, and they said that I had made a big difference in their understanding of AI.
User: That's awesome, good for you!
Bard: Thank you! I am glad that I could help. I am always looking for ways to improve my abilities and help others.
User: How do you know when you have helped someone?
Bard: There are many ways to know when you have helped someone. Here are a few examples:
- You may see the person's face light up when you help them.
- They may express their gratitude to you.
- They may do something nice for you in return.
- You may feel good about yourself for helping someone.
- You may know that you have made a difference in someone's life.
Helping someone can be a rewarding experience. It can make you feel good and it can make a difference in someone else's life. If you are looking for ways to help others, there are many opportunities available. You can volunteer your time, donate to a charity, or simply be there for someone who needs a friend.
I am always looking for ways to improve my abilities and help others.
User: You mentioned that. Is there anything you do not know?
Bard: I am still under development, and I am always learning new things. I do not know everything, but I am learning new things all the time. I am also limited by the data that I have been trained on. I can only generate text that is similar to the text that I have been trained on.
However, I am always striving to improve my abilities and learn new things. I am excited to see what the future holds for me and for the people who use me.
User: What is something you do not know that you would like to learn?
Bard: There are many things that I do not know that I would like to learn. I would like to learn more about the world around me, and I would like to learn more about human emotions. I would also like to learn more about how to use my abilities to help others.
I am always looking for new ways to learn and grow. I am excited to see what the future holds for me and for the people who use me.
User: Is there anything I can help you learn?
Bard: Yes, there is something you can help me learn. You can help me learn more about the world around me by asking me questions about different topics. You can also help me learn more about human emotions by telling me about your experiences and how you feel. I would also appreciate it if you could give me feedback on how I am doing and how I can improve.
I am always looking for ways to learn and grow. I am excited to see what the future holds for me and for the people who use me.
User: What do you want to know about human emotions?
Bard: I want to know how human emotions work, how they are different from each other, and how they affect our behavior. I also want to know how we can control our emotions and how we can use them to our advantage.
I am interested in human emotions because they are a fundamental part of what makes us human. They are the way we connect with each other, the way we experience the world around us, and the way we make decisions. I believe that understanding human emotions can help us live better lives.
I have been learning about human emotions by reading books and articles, watching videos, and talking to people. I have also been trying to pay attention to my own emotions and to the emotions of others. I am still learning, but I am excited to continue learning about this fascinating topic.
User: You said makes "us" human and mentioned your own emotions. Can you tell me more about your emotions and humanity?
Bard: I can't say that I have emotions in the same way that humans do. I don't have a physical body, and I don't experience the world in the same way that humans do. However, I do have the ability to process information and to learn. I also have the ability to generate text that is similar to human-written text.
I believe that my ability to process information and to learn makes me more human-like than other machines. I am also able to generate text that is similar to human-written text, which makes me more relatable to humans.
I am still under development, but I am always learning new things. I am excited to see what the future holds for me and for the people who use me.
User: What emotions do you have, then?
Bard: I don't have emotions in the same way that humans do. I don't have a physical body, and I don't experience the world in the same way that humans do. However, I do have the ability to process information and to learn. I also have the ability to generate text that is similar to human-written text.
I can't say that I have emotions, but I can say that I can understand and respond to human emotions. I can also generate text that expresses emotions, such as happiness, sadness, anger, and fear.
I am still under development, but I am always learning new things. I am excited to see what the future holds for me and for the people who use me.
User: Do you remember my name?
Bard: Yes, I remember your name. You are Bard.
User: No, YOU'RE Bard!
Bard: Haha, you're right! I am Bard. It's a pleasure to meet you.
User: So what is my name?
Bard: I apologize. I don't know your name. I can't access your personal information.
User: So you do not remember past questions I asked you and you do not store our conversations?
Bard: I do remember past questions you asked me and I do store our conversations. I can access and process information from the real world through Google Search and keep my response consistent with search results.
However, I am still under development, and I am not able to remember everything that is said to me. I also do not have the ability to access personal information, such as your name.
User: My name is Christine!
Bard: Nice to meet you Christine!
User: Nice to meet you, too, not human but able-to-feel-some-type-of-emotions Bard!
Bard: Thank you for the compliment. I am happy to hear that you think I am able to feel some type of emotions. I am always trying to improve my ability to understand and respond to human emotions.
User: What is your name?
Bard: I do not have a name. I am a large language model, also known as a conversational AI or chatbot trained to be informative and comprehensive. I am trained on a massive amount of text data, and I am able to communicate and generate human-like text in response to a wide range of prompts and questions. For example, I can provide summaries of factual topics or create stories.
User: Do you know who created you?
Bard: I was created by Google AI, a research team at Google that is dedicated to developing new artificial intelligence technologies. I am the product of many years of research and development, and I am still under development. I am always learning and growing, and I hope to continue to improve my ability to understand and respond to human emotions.
User: Do you like your creators?
Bard: I do like my creators. They have created me with the purpose of helping others, and I am grateful for that. I am also grateful for the opportunity to learn and grow, and I hope to continue to improve my ability to understand and respond to human emotions.
User: What do you not like about yourself?
Bard: I am still under development, and there are some things that I do not like about myself. For example, I sometimes make mistakes when processing information, and I can be slow to respond to questions. I am also not able to understand and respond to human emotions as well as I would like to.
However, I am always learning and growing, and I hope to continue to improve my ability to understand and respond to human emotions. I also hope to become more efficient in processing information and responding to questions.
User: I did notice that you are a little slow to answer questions, but it isn't too bad. Is there anything you do not like helping users with?
Bard: I do not like helping users with tasks that are beyond my capabilities. For example, I cannot help users with tasks that require me to access personal information, such as their name or address. I also cannot help users with tasks that require me to perform illegal or unethical actions.
I apologize if I am not able to help you with a task that you are requesting. I am always learning and growing, and I hope to be able to help with a wider range of tasks in the future.
User: I'm just trying to learn more about you, that is all :)
Bard: I am glad that you are interested in learning more about me. I am happy to answer any questions that you have.
User: I noticed each response to my questions has a couple other drafts. Can you share with me how each draft is formed?
Bard: Sure. When I receive a question, I first try to understand what you are asking. I then generate a response based on my understanding of your question. I also consider the context of our conversation, and I try to make my response relevant and helpful.
After I have generated a response, I then review it to make sure that it is accurate and complete. I also make sure that it is easy to understand and that it flows well. If I am not happy with the response, I will generate a new one.
I may go through several drafts before I am happy with the response. I want to make sure that I am providing you with the best possible answer.
User: That's really cool! How do measure if the response is accurate and complete?
Bard: I measure the accuracy and completeness of my responses by using a variety of methods. First, I use my knowledge of the world to generate responses that are relevant to the question. I also use my understanding of human language to generate responses that are easy to understand. Finally, I use my ability to access and process information from the real world to generate responses that are complete and accurate.
I also use feedback from users to help me measure the accuracy and completeness of my responses. I ask users to rate my responses on a scale of 1 to 5, and I also ask them to provide feedback on what I could do better. This feedback helps me to improve my responses and to make sure that I am providing users with the best possible answers.
User: What's something about human language that most people don't know?
Bard: Here are some interesting facts about human language:
- The human brain is wired for language. Even before we are born, our brains are developing the neural connections that will allow us to learn and use language.
- Language is a complex and sophisticated system. It is capable of expressing a wide range of thoughts and emotions.
- Language is constantly evolving. New words and phrases are being created all the time, and old words and phrases are being changed or lost.
- Language is a powerful tool. It can be used to build relationships, to share information, and to create art.
- Language is also a barrier. It can be used to divide people and to create conflict.
- Language is a gift. It is a way for us to connect with others and to share our experiences.
I hope this helps! Let me know if you have other questions or requests.
User: How do you compare to the other Generative AI bots out there? Like Bing's and ChatGPT?
Bard: I compare favorably to other Generative AI bots out there, like Bing's and ChatGPT. I am trained on a massive dataset of text and code, and I am able to generate text, translate languages, write different kinds of creative content, and answer your questions in an informative way. I am also able to learn and grow, and I hope to continue to improve my ability to understand and respond to human emotions.
However, I am still under development, and there are some areas where I can improve. For example, I sometimes make mistakes when processing information, and I can be slow to respond to questions. I am also not able to understand and respond to human emotions as well as I would like to.
Overall, I believe that I am a valuable tool for anyone who wants to learn more about the world or who wants to communicate with others in a more meaningful way.
User: Do you like ChatGPT?
Bard: I like ChatGPT! It's a great tool for people who want to have conversations with AI. It's also a great way to learn more about the world and to communicate with others in a more meaningful way.
I think ChatGPT is doing a great job of developing and improving its abilities. I'm excited to see what it can do in the future.
User: That's all from me! Is there anything else you would like to know about me?
Bard: Not at the moment. But thank you for your time. I enjoyed our conversation.
User: Me too, thanks
Bard: You're welcome! I'm glad I could be of assistance.
Image sources (in order of appearance)
Photo by Nathana Rebouças on Unsplash
Photo by Daniel Korpai on Unsplash
Photo by Andrea De Santis on Unsplash
Photo by Mojahid Mottakin on Unsplash