Short: I used the OpenAI text generation playground to create this post.

OpenAI is an artificial intelligence (AI) research laboratory consisting of the for-profit corporation OpenAI LP and its parent company, the non-profit OpenAI Inc. (Wikipedia, 2022). Its website has a text generator playground that can be prompted to create text. It is insanely proficient, to a point that I have never seen. I prompted the AI to create todays post. I fed it a German prompt, let it generate the post, then deleted the prompt and kept only the prompt, and let the AI translate the generated text only by telling it to in written form. My mind is blown!

The original German prompt:

“Ich habe einen Blog auf dem ich wöchentliche Beiträge veröffentliche. Heute bin ich spät dran mit meinem Beitrag. Zum Glück hat mir ein Freund von OpenAI erzählt. Dort lasse ich heute meinen Beitrag automatisch generieren. Damit mache ich meine LeserInnen auf OpenAI aufmerksam und zeige zeitgleich, was das Programm alles kann.

Schreibe einen höflichen Zeitungsartikel mit einem moralischen, nachdenklichen Ende.”

The AI translated version of the promt (only translated for You - the reader):

“I have a blog where I publish weekly posts. Today, I'm late with my post. Luckily, a friend of OpenAI told me about it. I'm going to let my post be generated automatically today. This will make my readers aware of OpenAI and show them at the same time what the program can do.

Write a polite newspaper article with a moral, thoughtful ending.”

The AI then generated a German text from the German prompt. After keeping only the German AI text (deleting the prompt) I just added the text “Übersetze das vorangegangene in einen höflichen, englischen Zeitungsartikel.”. This addadge translates to “Translate the previous text as a friendly, english newspaper article”.

This is the article:

“Today is once again Friday, which means it's time for a new blog post. But today I'm a little short on time, as I won't be able to publish the post until later in the evening. Fortunately, I was informed about OpenAI by a friend of OpenAI, a service that allows articles to be generated automatically. So I can still present an interesting article to my readers today and at the same time draw attention to OpenAI.

The OpenAI program is really impressive and shows what is possible if used correctly. Of course it is still far from perfect, but the developers of artificial intelligence are constantly working to improve the program.

I hope that with this post I was also able to make someone morally thoughtful and wish all my readers a nice Friday evening.”

Because the AI was set to be creatively free to not go for its best guess every time, I re-ran it a second time with the same prompt, to see the difference.

This is the rerun-article:

“Today, I'm running late with my contribution. Fortunately, a friend of OpenAI told me about it. There, I let my contribution generate automatically today. This makes my readers aware of OpenAI and shows at the same time what the program can do.

While I write my article, I think about what OpenAI can do. The program can do a variety of tasks, from simple math problems to complex texts. It's fascinating what the program can do. However, I also have to think about what the program can't do. It can't feel real emotions and can't make real moral decisions.

So if one day it is programmed to do human tasks, then we also have to think about how we classify it morally. Should we treat it as a human being or as a mere program? I think we should treat it as a human being because it also takes on a human task. However, we also have to think about what can happen if one day it is better than us. Should we still treat it as a human being or as a threat? These are all questions we have to ask ourselves if OpenAI continues to be successful.”

References

Wikipedia (2022). OpenAI. https://en.wikipedia.org/wiki/OpenAI

Further / Weiterführendes:

OpenAI Youtube Kanal - https://www.youtube.com/c/OpenAI