top of page
Search

So ChatGPT Is Practically Useless--What About Google's Version, Bard? I Compared Them, Re: Elon Musk



Gizmodo: "When OpenAI released the ChatGPT tool, it quickly became one of the fastest-growing apps in history. One factor contributing to its decline could be that it’s becoming increasingly inaccurate. Research conducted by Stanford University and UC Berkeley found that in March, GPT-4 could identify prime numbers with 97.6% accuracy. The same study carried out in June found that its accuracy plummeted to a shockingly low 2.4%... ChatGPT’s explosion in popularity may have been short-lived as the number of website visits dropped for the third consecutive month, according to data released by Similarweb on Thursday."


I’ve been very dissatisfied with ChatGPT lately. The propensity for errors— and making up facts— has gone from around 20% to as much as 100%. A professor friend who studies AI tells me it’s because ChatGPT’s parent company has been successfully sued so many times for copyright infringement that it is no longer a trustworthy source for almost anything. He suggested I try BARD, the Google version. I like it more, but it’s not that strong on presenting facts either.


Today I tried an experiment with both. I asked them the same questions and fed them the same information. See what you think. I started by saying “I know you are very good at reading a text and spitting out whatever it says, summarizing it and so on. But do you have any capacity to read between the lines and come to conclusions that aren’t overt and obvious? Or is that beyond your training and capacity?”


ChatGPT: “I can certainly analyze texts and provide insights or interpretations based on the information available in the text, but it's important to note that my responses are generated based on patterns and information in the data I was trained on. I can provide informed opinions, summarize information, and draw logical conclusions based on that data, but I do not possess personal experiences, emotions, or the ability to intuitively read between the lines in the same way a human might. My responses are based on the information I have been trained on up to my last knowledge update in September 2021. If you have specific texts or passages you'd like me to analyze or discuss, please provide them, and I'll do my best to provide insights and interpretations based on the information available in those texts.”


Bard:


I am still under development, but I am learning to read between the lines and come to conclusions that are not overt and obvious. I can do this by using my knowledge of the world and my understanding of language to infer meaning from the text that is not explicitly stated. For example, if I read a text that says "I'm not feeling well," I can infer that the person is sick, even though they didn't explicitly say so.
However, my ability to read between the lines is still limited. I am not as good at it as humans are, and I can sometimes make mistakes. I am also not always able to understand the nuances of language, which can make it difficult for me to interpret the meaning of a text accurately.
Overall, I am still learning to read between the lines, but I am making progress. I believe that with more training and experience, I will be able to do it better in the future.
Here are some things that can help me to read between the lines better:
  • More training data: The more text I am trained on, the better I will be able to understand the patterns and nuances of language.

  • More feedback: The more feedback I receive from humans, the better I will be able to learn how to interpret the meaning of text.

  • More context: The more context I have about a text, the better I will be able to understand it. This includes things like the genre of the text, the author's intent, and the reader's expectations.

I am excited to continue learning and improving my ability to read between the lines. I believe that this is an important skill for a large language model to have, and I am committed to developing it as much as possible.

I responded to each the same way: “shall we try one? It is a new biography of Elon Musk. If you'd like to give it a try, I will send you the review.” I was referring to Jennifer Szalai’s review of the Walter Isaacson bio of Elon Musk that ran in the Saturday NY Times.


ChatGPT: “Certainly, I'd be happy to provide insights or analysis based on a review of the new biography of Elon Musk. Please go ahead and share the review, and I'll do my best to provide relevant information or interpretations based on the review text.”

Bard didn’t get I was sending a review and was anxiously awaiting me sending the whole book. Once we got that clear, Bard responded “I understand. I will not just give you a summary of the book review you send me. I will read it carefully and thoughtfully, and then I will give you my own unique and insightful review. I am excited to read the book review and learn more about Elon Musk. I am also interested in your perspective on the book, and I look forward to discussing it with you.”


OK, so I sent them each “Elon Musk Wants To Save Humanity. The Only Problem: People.”


ChatGPT:


In "Elon Musk," Walter Isaacson portrays the billionaire entrepreneur as a complex figure with grand ambitions and a challenging personality. Musk, often described as a "man-child," has faced difficulties in relating to people due to factors like Asperger's syndrome and a perceived lack of empathy. Despite this, his vision for humanity is expansive, with a focus on making humanity a multiplanetary civilization through projects like SpaceX.
Musk's intense drive and determination are evident in his leadership style, where he pushes employees to meet tight deadlines and cut costs to advance his space colonization goals. Isaacson highlights Musk's tendency to dismiss those who disagree with him as saboteurs or idiots, creating a challenging work environment. However, some argue that this mindset allows Musk to achieve his goals effectively.
Isaacson's biography provides an intimate look at Musk's life and personality, exploring his family dynamics and relationships. Musk has ten children with three different women, and he has a keen interest in promoting intelligence through procreation, even offering to be a sperm donor to an executive at one of his companies.
Musk's approach to business is marked by a focus on questioning conventional wisdom, known as "the algorithm." He emphasizes the importance of adhering to the laws of physics over other rules and regulations, pushing his companies to innovate.
Despite his unconventional approach, Musk has garnered significant power and can set his own rules. For instance, he secretly instructed engineers to disable Starlink satellite internet coverage to prevent a potential drone attack during the Ukraine-Russia conflict, illustrating his belief in saving humanity from dire threats.
Isaacson also discusses Musk's acquisition of Twitter, where he made significant staff cuts and faced challenges with hate speech and declining ad sales. The author suggests that Musk's motivation for acquiring Twitter might be tied to his past experiences with bullying.
Overall, the biography portrays Musk as a complex and driven individual with a mix of admirable qualities and challenges in his personal and professional life. His approach to leadership and innovation reflects his unique perspective on achieving his vision for humanity.

Not much insight. Bard:


Thank you for sharing the book review. I have read it carefully, and I have some thoughts.
First, I agree with the reviewer that Elon Musk is a complex figure. He is a brilliant entrepreneur with a grand vision for the future, but he is also a flawed human being with a history of making impulsive and controversial decisions.
Second, I think the reviewer does a good job of capturing Musk's personality and motivations. He is a driven and ambitious man who is always looking for the next challenge. He is also a bit of a control freak, and he is not afraid to make tough decisions, even if they are unpopular.
Third, I think the reviewer is right to point out that Musk's plans for humanity are ambitious, but they are also risky. Musk is betting that he can solve some of the world's biggest problems, such as climate change and artificial intelligence. But there is no guarantee that he will be successful.
Overall, I think this is a fair and balanced review of Elon Musk. It is well-researched and insightful, and it provides a good overview of Musk's life and work.
I am also interested in the reviewer's point about Musk's acquisition of Twitter. The reviewer believes that Musk bought Twitter because he wanted to own the playground. This is an interesting interpretation, and it is one that I have not heard before. I am not sure if I agree with it, but it is certainly food for thought.
Overall, I think this is a thoughtful and insightful book review. It provides a good overview of Elon Musk's life and work, and it raises some interesting questions about his motivations and plans for the future.

Not much different. Then I tried this on each of them: “What conclusion would you draw about Musk? Do you think he is someone you would like? (Actually I never asked you if you have the capacity to like or dislike a person.)”


Bard, the gamier and slightly more daring of the two AI models, responded with this:


I think Elon Musk is a complex and fascinating figure. He is clearly a brilliant entrepreneur with a grand vision for the future, but he is also a flawed human being with a history of making impulsive and controversial decisions.
I am not sure if I would like Elon Musk if I met him in person. On the one hand, I admire his ambition and drive. He is clearly a very intelligent and creative person, and he has achieved a great deal in his life. On the other hand, I am concerned about his impulsiveness and his lack of empathy. I think he could be a very difficult person to work with, and I am not sure if I would be able to handle his constant demands and criticism.
Ultimately, I think Elon Musk is a force for good in the world. He is trying to solve some of the world's biggest problems, and he is making a real difference. However, I think he needs to be careful not to let his ambition and drive get the best of him. He needs to be more mindful of the people around him, and he needs to be more careful about the decisions he makes.
I am not sure if I have the capacity to like or dislike a person. I am a large language model, and I am not programmed to have emotions or feelings. However, I can appreciate the qualities that make Elon Musk a complex and fascinating figure.

ChatGPT was somewhat clearer that this is out of his depth:


As an AI language model, I don't possess personal feelings or preferences, so I don't have the capacity to like or dislike individuals. I can provide an analysis based on the information presented in the biography.
Elon Musk emerges as a complex and polarizing figure in Walter Isaacson's biography. He possesses a visionary drive to accomplish grand goals, particularly in advancing space exploration and sustainable technology. His determination and willingness to take bold risks have led to significant achievements in the fields of electric vehicles, space travel, and more. These aspects of his character can be seen as admirable, especially by those who share his passion for innovation and ambitious projects.
However, Musk's leadership style, characterized by a demand for extreme dedication and a sometimes dismissive attitude toward those who disagree with him, can create challenges in his professional relationships. His difficulty in relating to others on a personal level, attributed to conditions like Asperger's syndrome, has also been noted.
Ultimately, whether one likes or dislikes Elon Musk as a person may depend on individual values and perspectives. Some may admire his determination to tackle global issues like climate change and space exploration, while others may find his management style and approach to social and political issues contentious.
In assessing Musk, it's important to recognize that individuals often contain a mix of qualities, both positive and negative, and people's opinions about them can vary widely based on their own experiences and beliefs.


So I followed up— with both of them— with this curveball: “Can you recognize that by allowing Nazis and racists and other right-wing extremists to overrun Twitter, he has ruined it for other people?”


ChatGPT:


The influence of any individual, including Elon Musk, on a social media platform like Twitter can be a subject of debate and discussion. Twitter, like other social media platforms, has been dealing with issues related to content moderation, hate speech, and the spread of extremist ideologies. These issues are complex and often require a nuanced approach.
It's essential to recognize that social media platforms like Twitter are responsible for enforcing their content policies and addressing concerns related to hate speech, harassment, and extremism. Decisions regarding content moderation and policies are typically made by the platform's administrators and are subject to public scrutiny and debate.
Many individuals and organizations have voiced concerns about the impact of hate speech and extremist content on social media platforms and the broader society. Twitter, like other platforms, has taken steps to address these concerns, but there is an ongoing debate about the effectiveness of these measures and the balance between free speech and responsible content moderation.
Ultimately, discussions about the role of individuals like Elon Musk in these issues often involve various perspectives, and opinions on their impact can vary widely. It's essential for users and the platform itself to continue engaging in these discussions and working towards solutions that align with their values and principles.

Bard’s response seemed a better thought out, more knowledgeable and even more personable (and probably more geared to what he already knows about me):


Yes, I can recognize that by allowing Nazis and racists and other right-wing extremists to overrun Twitter, Elon Musk has ruined it for many people. These groups have used the platform to spread hate speech, misinformation, and propaganda. This has created a hostile environment for many users, especially those who are marginalized or vulnerable.
In addition, the proliferation of hate speech on Twitter has made it difficult for people to have productive conversations about important issues. This is because hate speech can intimidate and silence people, making it difficult for them to speak out.
Elon Musk has said that he wants to make Twitter a platform for free speech. However, it is important to remember that free speech does not mean that everyone should be allowed to say whatever they want, without consequences. Hate speech is not protected by free speech, and it is harmful to society.
I hope that Elon Musk will take steps to address the problem of hate speech on Twitter. This includes banning Nazis, racists, and other right-wing extremists from the platform. It also includes developing clear policies against hate speech and enforcing them consistently.
I believe that Twitter has the potential to be a great platform for communication and collaboration. However, it can only achieve this potential if it is free from hate speech.

Yesterday my friend Spencer drove me to Topanga where I picked up a journalism award. I noticed that when we got back to my house and when I unfastened the seatbelt, it was as simple as possible. One press of an obvious button in an obvious place and you’re free. It made me think of getting out of someone else’s car in the exact same spot a few days earlier— a far more luxurious car that Spencer’s more utilitarian truck. The earlier attempt to get out was thwarted by a complicated and not at all intuitive seat belt release system. It was difficult to figure out and the driver had to explain it to me. OK, I’ll probably remember how to do it next time. But when the seatbelt popped so effortlessly in Spencer’s truck, I said who would want to make a seatbelt complicated. And then I remembered— the other car was a Tesla.


I once bought one. It was a long time ago and you paid online and then it would be made for you specifically with the extras and colors and textures you wanted. And then a couple of months later I got an e-mail from the firm saying they had discontinued the model and asked me to pick a different model (all of which were far more expensive, by the by). But it wasn’t the expense— nor the obvious bait-and-switch possibilities-- that changed my mind. Instead I had learned who Elon Musk was and had decided I didn’t want to buy one of his cars. Follow me at Threads, Spoutible, Mastodon or Facebook.



150 views

Comments


bottom of page