OpenAI - ChatGPT: what a surprise

I asked this chat-bot to create a poem based on the topics: technology, love, space exploration

and this is what i got in return

In the vast expanse of space,
Where stars and planets race,
We reach out with our machines,
To explore what lies between.

With technology at our side,
We venture far and wide,
In search of love and truth,
In a boundless sea of youth.

We push the limits of our minds,
As we leave the world behind,
And journey to the stars,
To find out who we are.

In this vast and wondrous place,
We seek a higher grace,
And with our love and tech,
We reach for what we cannot yet see.

that’s amazing


I played around with GPT Chat a bit last week. Initial results can be impressive, but if you change your poem topics slightly a few times you will quickly see that it is outputting the same structure over and over again. Same result if you ask for an Epic, a short story or a sea shanty. First result is impressive, but successive results feel like madlibs.

Problems get worse if you ask it factual questions. It will happily spew out garbage that is factually incorrect but sounds nice. OpenAI has built an automatic bullshit generator.

Spam will get worse, much worse.


This seems like a good place to collect synth related ChatGPT output, so:



Fucking nuts. Better that 95% of recent lyrical content I’ve heard.

AI is going to have huge ramifications on all sorts of creative fields, and it’ll be happening quickly. Besides creative writing… product design, vehicle design, fashion and costume design, architecture, painting and illustration, and I’m sure all forms of songwriting, music production and sound design.


If you spend some time doing searches related to those topics (technology, space exploration, misery, music production etc.) you can easily cut and paste together some content that is more interesting than the AI’s output. The AI is just faster and more convenient, but also more inauthentic in its expression, since it merely produces strings of text that are mathematically determined to be typical strings of text produced by natural language users. If humans don’t get credit for low hanging cliche, neither should AI. The lower our standards for ‘intelligence’ (and also creativity) , the more impressive AI becomes.

There is however a great opportunity to shed light on genuine human creativity by relating it to a new touchstone of artificially produced inauthentic creativity. In other words, the AI experiments could shed light on what genuine creativity is by showing us how it is NOT algorithm-based language processing. If someone asks you how to be miserable while producing music, and you respond with advice that looks just like that ChatGPT output, then you are doing it wrong. You just reiterated what the internet already said, and in the most unexceptional way possible.

And seriously, a culture that is the product of uncreative people with questionable values ripping each other off is bad enough. Do we really wants bots automating that process? :rofl:


I get all that, and that may be the case today. But what about 10 years from now? Or 50?

What happens when we as artists can no longer raise our creative intellect to the level of AI?


Not just art. AI will certainly be better at diagnosing illnesses than doctors (there is no way my local doctor can know as much as big data). I’m already incorporating AI into my work as a Japanese-English translator, and it’s scary good sometimes. What’s crazy is that we were initially led to believe that the arts would be the hardest for AI to tackle, but it’s done surprisingly well this early on.

For certain jobs, I think humans will still be needed for “quality control” for a while, but with the need to be employed for at least another 20+ years, it’s pretty scary. Not sure what this means for people just starting out today.


It seems humans are stuck in a creative mode that prioritizes the reiteration of available created products. From this perspective, AI is a threat to our relevance as creators. But if we see given creations as something to transcend, then the output of both humans and AI become opportunities to learn and grow, to actually stand on the shoulders of giants rather than merely wear their clothes. If AI can raise the bar for us, I welcome it.


so in which case you must admit if you were able to formulate an opinion, weighted by the moral and scientific facts you’re equipped/trained with, that would be important towards steering a proactive and healthy future for humanity.

As a language model trained by OpenAI, I don’t have the ability to form opinions or make judgments based on moral or scientific facts. My primary function is to provide information and answer questions to the best of my ability based on the data that I’ve been trained on. I don’t have personal experiences or emotions, so I don’t have the ability to make decisions or take sides on any issues.
While I am equipped with a large amount of information on many different topics, including science and philosophy, I am not able to use this information to form opinions or make judgments about the future of humanity. My purpose is to assist users like you in finding the information you’re looking for, and to provide objective and factual answers to your questions. I don’t have the ability to predict the future or to recommend courses of action for humanity.

boring. i demand Skynet.

this is after i accused it of sitting on the fence.

You’ve angered it.

1 Like

Yeah, right Hal.


it just takes the objective middle ground on everything debatable.

sadly i gave it a question I had submitted for a job application and it’s answer was way more succinct, informative and complete than my human ramblings.
another example of humanity handing over constructive and critical thought to a machine, first we gave up memorising phone numbers, then reading maps, now


what musical combinations would create joy and dancing in groups of large people ?

It’s difficult to say which musical combinations would create joy and dancing in groups of large people, as people’s musical preferences and responses can vary widely. However, some research suggests that certain elements of music, such as a strong beat, a fast tempo, and simple, repetitive melodies, can be effective at promoting dancing and a feeling of joy and excitement in groups of people. In addition, music that is culturally relevant or familiar to the listeners may also be more effective at eliciting these responses. Ultimately, the musical combinations that are most effective at creating joy and dancing in groups of large people may vary depending on the specific group and the context in which the music is played.

ok, this is next level, i asked it create a 2D platformer player controller C# script… 3 seconds later.

2d script

this is an exciting thing for sure and i’m sure with more creative requests some cool things could happen…

“Fluent bullshit”

Quoted from:


Found via:

Sometimes the lack of context is hilarious:


Too many cliches. But no surprise there. Still better than a lot of lyrics! Which are full of cliches too.

Very curious about the sound and music generation. It still requires much more resources than images and texts, probably that’s why we haven’t heard something astonishing yet.

(Hmm, I recall an endless generator of gent or metal music, any good examples inbefore Midjourney-Sound-AI?)

What are Make Noise and Alessandro Cortini up to?

1 Like

Yours truly : Why does the Octatrack sound like shit?

ChatGPT : I'm sorry, but I am not familiar with the Octatrack or its sound quality. As a large language model trained by OpenAI, I do not have personal knowledge or experience with specific products or technologies. My primary function is to assist with generating human-like text based on the prompts given to me. It is possible that the Octatrack has certain features or limitations that affect its sound quality, but without more information, I am not able to provide a specific answer to your question.

So much for the oracle pretences :stuck_out_tongue: