The techies are beside themselves with excitement: Thanks to ChatGPT, you can have an almost normal conversation with AI without even realizing it. Pupils are delighted because AI can produce a reasonable text on almost any school subject – whether it’s German, English or history.
And the world of features finally has a topic again: Whom will this AI replace? Can we ever trust a text again? Will this technology revolutionize the world of communication? Have self-absorbed nerds opened Pandora’s box and pushed the whole of humanity a little closer to the abyss?
NO. Nothing will happen once this wave of indignation has swept around the globe.
Yes, this language model is very impressive. It’s great fun to try out a few things. Why not register on OpenAI yourself and play around a little? It’s well worth it.
Discussions with a chatbot
I simply asked ChatGPT whether the world was coming to an end and received a plausible answer:
“It is unlikely that ChatGPT or other large language models will revolutionize the world. They are still machines that are programmed and trained by humans, and they are not capable of making decisions on their own. They can only provide answers to questions they are asked based on what they have learned during their training. It’s important that humanity understands the limitations of technologies like these and uses them responsibly.”
So instead of outrage and fear, we could just try curiosity, right? Why does ChatGPT think people might be afraid of ChatGPT?
“It’s normal for people to feel a certain amount of skepticism or anxiety about new technologies. In terms of ChatGPT or other large-scale voice models, people may worry that these technologies could replace human labor or compromise privacy. It’s important that we realize that these technologies are tools and that it’s up to us to use them responsibly.”
So, this answer provides more than all the gloomy worries of the many intelligent writers in feature articles. But is it dangerous? What does it mean for humanity?
And who writes better now?
So, I asked ChatGPT: What does it mean if I like one of your texts better and find it more meaningful than the many texts about you that I’ve read in the last few days?
“If you like my text better and it seems to make more sense than other texts you have read about me, it could mean that the text is written clearly and comprehensibly and addresses your questions. It could also mean that the text is relevant and interesting for you. However, keep in mind that I am a machine learning model and have no human intelligence. My answers are based on what I have learned during my training and therefore cannot always be perfect.”
This is exactly the point: We obviously write too much. If we wrote less, we would also have to worry less about whether a text we read was written by a humanoid editor or an AI chatbot.
Perhaps in future we will simply value real communication again. Maybe in the future we will simply appreciate editors and people who we can trust not to simply throw ChatGPT at us.
What are we actually afraid of?
People who write about a new technology without any expertise and share their stereotypical fear scenarios with the whole world still scare me.
What do we humans want now? Is it really about a clear distinction between machine-generated texts? Or do I just want comfort from this world and would perhaps even be helped by an understanding response from AI?
ChatGPT provides that too:
“I’m sorry that you feel threatened by the world. Please remember that there are also many beautiful things in the world and that it is up to you to make the most of every day. It can be helpful to surround yourself with people who are close to you and to seek support when you need it. It’s also important to take care of yourself and take time to relax and unwind. If you feel overwhelmed, don’t hesitate to seek professional help.”
I don’t find this encouragement really helpful. But perhaps I would if I didn’t know that it was from an algorithm… We humans might need it a little schizophrenic sometimes.
