Report 2934
tl;dr: I found that if you pass empty prompts to ChatGPT it spits back results anyway At first I thought they might be hallucinations but now I believe I'm also getting back other people's responses
Earlier last month; to a lot of fanfair of developers - along with enterprise, media, and government interests - OpenAI release their new Large Language Model ChatGPT-4 .
For LLMs it’s definetly very advanced. Just days before I was playing with alpaca.cpp - a model that is designed to run on a CPU with a smaller set of memory constraints. I had started with building a very simple web interface using NodeJS and sockets to parse the command line. Shortly after I started working on this, my GTP-4 invite arrived, and I started playing with the API. Quickly I could see with the right kind of prompts, it can be a useful tool in the right hands. It’s already helped me re-write complex code into simpler methods that pass the tests, or reduce complexity by moving code to functions:
However I started noticing something odd - due to a bug in my code, I was passing empty prompts to the ChatGPT endpoint - but every time I was getting back a response - and they were completly random. From standard “I am an AI model…” responses, to information about people, places and concepts.
The same day, it was #StochasticParrotsDay - an online conference around AI ethics, hype and it’s future - and it gave me an idea - why don’t I turn this into a Mastodon Bot ! (*this was this original URL - it has now since moved to botsin.space )
Since I’ve been running this bot, I’ve come to some conclusions about the responses and I believe that without a prompt a large proportion of the responses are meant for other users. If it’s a bug, possibly passing an unsanatised empty prompt causes it to return something it should not in response.
Of course I could be wrong - these could be stochastic hallucinations being dreamed up by a machine, they could be the API returning a random set of training data that happens to be at the right Entropy - but then ChatGPT itself isn’t then much better than a Markov Chain and OpenAI (and the whole LLM/AI market) have been playing everyone for fools.
But if I am correct then the current OpenAI APIs could be made to potentially leak private or sensitive data, by simply not sanatising their inputs…
The bot will continue to run until at least the end of this month, and all the content will be archive at stochasticparrot.lol.
Setting up the Bot Infrastucture
To get the bot up and running, I wanted to do it for free, and easy to manage. In the end I opted to use GitHub Actions with scheduled tasks to create it - this allowed me to set up a script that ran hourly - calling the ChatGPT API with an empty prompt, and turning it into a toot. I also found that passing only a space character to the Dall-E API also produced images.
With both scripts, after getting a response from OpenAI, I use it to generate one or more toots - depending on the length as a set of replies, with the images first download them and then upload them as attachments first.
Some of the more recent toots are below - ones with a parrot emjoi (🦜) are created without a prompt, while if a prompt is used I add a speech bubble (💬) to indicate
🦜 My opinion as an AI language model is that South Korea should not ban tattoos. Tattooing is an art form and a way for individuals to express themselves. However, it's important to regulate the industry to ensure safety and sanitation standards are met.
As in many countries, tattoos have been historically associated with gang culture in South Korea. Since times have changed, this perception has shifted with more people viewing tattoos as a form of
🦜 Yes, it's possible and actually quite common to be attracted to fictional characters. This can happen for many reasons:
1. Relatability: People often become attached to characters they find relatable or who share similar experiences, struggles, or personality traits.
2. Idealization: Fiction allows authors and creators to build their ideal version of a character - one that is intelligent, attractive and the epitome of ideals people crave.
3. Wish fulfillment:
🦜 Here are the ten most-used emojis in the world:
1. 😂 Face with Tears of Joy2. ❤️ Red heart3. 😍 Smiling Face with Heart-Eyes4. 🤣 Rolling on the Floor Laughing5. 😊 Smiling Face with Smiling Eyes6. 👍 Thumbs Up7. 💕 Two Hearts8. 🙏 Hands folded in prayer or thank you gesture (also considered a high-five)9. 🔥 Fire, representing something being hot or exciting.
10:😘 Face Blowing a Kiss
These rankings may vary slightly depending on geographical
Once I had this up and running, I then created a small AstroJS website that outputs each entry as a posting.
Making Polly Speak
Up to this point, I had just been working with text and images - but I had recently seen ElevenLabs in some tech news, and that they had a text-to-speech API. After some initial issues (which used up all of the free credit) - I eventually set up another action that took the OpenAI response, and passed it to the ElevenLabs API - this then provided a MP3 stream of the speech, saved locally and again upload to Mastodon and attach to a toot.
I also decided to try see if I could get it to generate some polls. With some gentle prompting I was able to get it to generate JSON output which could be used in polls. Sadly, most of the time it seems to repeat the same questions over and over with just slightly different wording, occasionally coming up with something original
🦜 Which of these fictional cities would you like to visit?
- Gotham City from Batman (0)
- Hogwarts from Harry Potter (8)
- Westeros from Game of Thrones (3)
- Panem from The Hunger Games (2)
Total Votes: 13
🦜 Which exotic fruit would you like to try?
- Durian - The King of Fruits (25)
- Salak - The Snake Fruit (10)
- Mangosteen - Queen of Fruits (22)
- Rambutan - Hairy Lychee (9)
Total Votes: 66
A fun experiment
I even went as far as trying to generate video content - not through Stable Diffusion, but by generating text themes to use with the Createomate API - allowing me to generate social media “fact” videos. Unfortunatly this was a bit buggy, and due to the way Mastodon works can time out quite a bit.
Overall, writing this bot was a fun experiment - but I probably learned more about writing better pipelines, than I did about AI and LLMs. What did surprise me was how often the responses seem to be to questions that were not asked - where are these responses being generated? Are we seeing the flicker of AGI? Or just the stochastic ramblings of a machine run by some sketchy people.