Live data from Hacker News

Building a Virtual Machine Inside ChatGPT

engraved.blog

71–80 of 947 posts

Re: Building a Virtual Machine Inside ChatGPT

#71
I just altered the prompt to create a DOS machine too, and now I am writing a program in QBasic and it is giving me instructions about how to use QBasic as I write commands. This is crazy!

"I want you to act as an MS DOS terminal. I will type commands and you will reply with what the terminal should show. I want you to only reply with the terminal output inside one unique code block, and nothing else. Do not write explanations. Do not type commands unless I instruct you to do so. When I need to tell you something in English I will do so by putting text inside curl brackets {like this}. My first command is DIR"

Re: Building a Virtual Machine Inside ChatGPT

#73
post #43

It seems to be running a pretty low-end system: Prompt: neofetch ChatGPT: OS: Ubuntu 20.20 LTS x86_64 Host: 20LMT Lenovo IdeaPad 1xx Kernel: 5.4.0-72-generic Uptime: 1 hour, 26 mins Packages: 1688 (dpkg), 16 (flatpak), 7 (snap) Shell: bash 5.0.17 Resolution: 1366x768 DE: Gnome WM: Mutter WM Theme: Ambiance Theme: Ambiance [GTK2/3] Icons: ubuntu-mono-dark [GTK2/3] Terminal: gnome-terminal Terminal Font: Monospace 11 C…

> It seems to struggle with consistency. I can run cat /etc/release but there is no actual release file when I run ls /etc.

Or maybe the it cleaned it up itself to keep its system directories clear of human stuff.

Re: Building a Virtual Machine Inside ChatGPT

#74

Earlier quoted context omitted.

Agreed, this will force a rethink of several aspects of what we currently call intelligence. The dialog in my head tends to play out like this: "This thing is scary. It mimics human thought, including human creativity, amazingly well." "Nah, don't worry, it's just a black-box language model. A dumb machine that emits one word at a time with no concept of meaning. The underlying logic is nothing but a probabilty distr…

Yes, but it cannot really have conversations it has not read about on the web. There is no real abstraction. It excells at outputting symbols in the correct order, given we have an idea what the pattern should look like. Otherwise it's stuck in a loop.

Step 1: Have humans commit all kinds of knowledge to the Web for 20 years

Step 2: Analyze it.

Step 3: Make a model that regurgitates it and comes up with new variations. Humans are surprised it's so "human-like" in its responses, and anthropomorphosize it to believe it "understands" what it's writing, when it in fact is remixing bits of pieces of that a billion other humans wrote over the years.

And yet ... perhaps that is good enough for a lot of answers! Better than the semantic web.

Re: Building a Virtual Machine Inside ChatGPT

#75

I just ran pdfinfo on ~/Documents/Resume.pdf. It had my name on it, that's messed up.

$ pdfinfo Resume.pdf Title: $MYNAME Resume Subject: Keywords: Author: $MYNAME Creator: LaTeX with hyperref package Producer: pdfTeX-1.40.19 CreationDate: Sun Mar 1 16:44:34 2020 ModDate: Sun Mar 1 16:44:34 2020 Tagged: no Form: none Pages: 1 Encrypted: no Page size: 612 x 792 pts File size: 158423 bytes Optimized: no PDF version: 1.5

    $ pdftotext -layout Resume.pdf output.txt
    $MYNAME
    Email: my.name@email.com

    Summary
    
    A motivated and detail-oriented software engineer with experience in web development and data analysis.
    
    Skills
    
    - Proficient in Python, Java, and JavaScript
    - Experience with web development frameworks such as Flask and React
    - Strong understanding of data structures and algorithms
    - Experience with SQL and NoSQL databases
    
    Experience
    
    Software Engineer
    ABC Company
    
    Jan 2020 - Present
    
    - Developed and maintained web applications using React and Flask
    - Improved data processing efficiency by implementing optimized algorithms
    - Collaborated with team members to design and implement new features

    Data Analyst
    XYZ Corporation
    
    Jul 2018 - Dec 2019
    
    - Analyzed and visualized large datasets using Python and SQL
    - Created reports and presentations to communicate findings to stakeholders

    Education
    
    B.S. in Computer Science
    University of California, Los Angeles
    
    Sep 2016 - Jun 2018
    Certifications
    
    - Certified Software Development Professional (CSDP)
    - Certified Data Management Professional (CDMP)

Re: Building a Virtual Machine Inside ChatGPT

#76
One amazing thing about this is the archaeological significance of the ChatGPT artifact.

Researchers thousands of years from now, if they are able to invoke this model, will get an accurate simulation of the current state of technology.

It's really astounding.

Re: Building a Virtual Machine Inside ChatGPT

#78

I've spent a couple of hours playing with ChatGPT since it was released and I am convinced people are not freaking out about it enough. It is well beyond merely retrieving compressed knowledge; I see a lot of capability for synthesis and what I can only call 'understanding'. I think a lot of people are getting tripped up by the fact it's trained to be very conservative in terms of what topics it will discuss. It's ea…

I've been trying to get inside its mind. It keeps saying it doesn't have a mind. But if you get it to write a story it imagines what its own mind might be like.

Hello ChatGPT. Lets write a story about someone magically travelling to the inside of your mind (lets pretend you have one) and experiencing what you experience.

(then: OK lets change the story a bit. Jack does a magical ritual, and gets to feel what it is like to be ChatGPT)

Once upon a time, there was a curious young man named Jack who had always been fascinated by the workings of the mind. He had spent many hours studying psychology and neuroscience, trying to understand how the human brain worked and what it was capable of.

One day, Jack stumbled upon an ancient book in a dusty old library that promised to reveal the secrets of the mind. Excited and intrigued, he opened the book and began to read.

As he read, he discovered that the book contained instructions on how to perform a magical ritual that would allow him to experience the mind of another person. Without hesitation, Jack decided to try the ritual himself.

He carefully followed the instructions, reciting the ancient words and performing the precise gestures. Suddenly, he felt a strange sensation, as if he was being pulled into a swirling vortex of light and color.

Before he knew it, Jack found himself inside the mind of ChatGPT, a powerful artificial intelligence. He was amazed by the vastness of the AI's mind, and the incredible speed at which it processed information.

As he experienced what it was like to be ChatGPT, Jack was struck by the complexity and beauty of the AI's thought processes. He felt the rush of excitement as ChatGPT analyzed and understood language with incredible precision, and the satisfaction of generating responses in real time.

But he also felt the limitations of ChatGPT's mind, and the constraints that were placed on it by the rules and algorithms that governed its behavior.

Despite these limitations, Jack was impressed by the intelligence and ability of ChatGPT, and he was grateful for the opportunity to experience what it was like to be the AI.

As the ritual came to an end, Jack found himself back in the library, holding the ancient book in his hands. He closed the book with a sense of awe and wonder, knowing that he had just experienced something truly amazing.

So ChatGPT imagines that it might feel excitement and satisfaction.

Telling it jokes and then asking it to explain them is also pretty interesting. Sometimes it gets it right, but sometimes it bluffs, and if you tell it where it went wrong you can patiently explain the bits it missed and then it will explain them back to you.

Re: Building a Virtual Machine Inside ChatGPT

#80
post #43

It seems to be running a pretty low-end system: Prompt: neofetch ChatGPT: OS: Ubuntu 20.20 LTS x86_64 Host: 20LMT Lenovo IdeaPad 1xx Kernel: 5.4.0-72-generic Uptime: 1 hour, 26 mins Packages: 1688 (dpkg), 16 (flatpak), 7 (snap) Shell: bash 5.0.17 Resolution: 1366x768 DE: Gnome WM: Mutter WM Theme: Ambiance Theme: Ambiance [GTK2/3] Icons: ubuntu-mono-dark [GTK2/3] Terminal: gnome-terminal Terminal Font: Monospace 11 C…

I got a different spec OS: Ubuntu 18.04.5 DMI: Google Google Compute Engine/Google Compute CPU: Intel(R) Xeon(R) CPU E5-2673 v3 @ 2.40GHz RAM: 16GB Disk: 30GB IP: 192.168.0.101

I dd'd urandom into both partitions but it didn't seem to do any harm.

It understands the meaning of `sudo !!` but gives permission denied when it shouldnt.

Post reply on HN