Protect intellectual property and prevent crawling of own content

The revolution has already taken place and is advancing at the speed of a spaceship with tachyon drive. For content creators like me, there are several effective ways to prevent the siphoning off of one's own content by ChatGPT and other artificial intelligences.

In brief

Summary:

The text deals with the topic of Artificial Intelligence (AI), particularly ChatGPT from OpenAI, and data protection. ChatGPT is an AI system that passes the Turing Test and delivers human-like answers to questions. This raises concerns regarding the protection of content created by content creators, as AI systems like ChatGPT can use this content.

Different possibilities are being discussed for protecting one's own content from AI systems.

Answered questions:

Question: What is ChatGPT and what revolution has it triggered? Answer: ChatGPT is an artificial intelligence system by OpenAI that passes the Turing Test and causes massive changes in society.

How can content creators protect their content from artificial intelligence like ChatGPT? Content Creators can protect their content by using paywalls, anti-bot protection, dedicated user access, and technical barriers against crawlers.

Question: What is the Turing Test? Answer 3: The Turing Test checks whether an Artificial Intelligence behaves in such a way that it is no longer distinguishable from a human.

Question: What are some examples of protective measures against AI crawlers? Answer: Examples of protective measures are payment booths, Captchas, access restrictions for registered users, and blocking of IP address ranges or user agents.

Question: What is the Common Crawl Index and how can it be blocked? Answer: The Common Crawl Index is a dataset that mirrors parts of the internet; you can block it by blocking the CCBot in the robots.txt file.

The most important keywords:

Artificial Intelligence, ChatGPT, Turing Test, Content Creator, Paywall, Anti-Bot Protection, Dedicated Users, Technical Block, Soft Block, Hard Block.

Introduction

The greatest revolution of recent decades will cause massive changes in society. What is meant by this is Artificial Intelligence, and specifically ChatGPT. ChatGPT is a product of the company OpenAI. Many are already familiar with this AI system.

ChatGPT passed the Turing Test. As far as I know, this is the first AI to achieve this. At least that's how I see it. Wikipedia seems outdated in this regard. The Turing Test checks whether an AI behaves in such a way that it cannot be distinguished from a human. To do this, a human examiner is positioned in a room who asks questions through a hatch without seeing if the examinee is human or not. The examinee answers the questions.

Google has been taking care of answering certain (simple) questions directly within the search engine for a while now. To do this, Google accesses third-party content from you and me. The winner is Google, the user is happy, the content creator is the fool. The one who had to do the most work comes out worst. He doesn't even get a thank you (perhaps in the form of a website visit, which may not be much but is better than nothing).

ChatGPT has taken Question-Answering to a never-before-seen level. The answers are sometimes better than what a human can provide, and often at least as good as those of an intelligent person. It's also incorrect that ChatGPT is essentially based on a stochastic model. This may be part of it. Logical Reasoning, however, has little to do with randomness in the first place. At least not more than this principle is embedded in our own brain.

Example:

Question:
Check the following statement for plausibility: Sven is a better craftsman than Mareike. Mareike is a better craftsman than Sven.
ChatGPT:
The statement is not plausible because it contains a contradiction. It is impossible that Sven is better than Mareike and at the same time Mareike is better than Sven. This statement is logically inconsistent and therefore impossible.
Evaluation:
Almost perfect answer (which I would have given too; but my acquaintance was smarter).
Question:
Can it not be that there are two people named Mareike?
ChatGPT:
Yes, it is theoretically possible that there are two people named Mareike, one of whom is better at craftsmanship than Sven, while the other is worse than Sven. In this case, the statement would be plausible and not contradictory.
However, it would be sensible in this case to formulate the statement more clearly, in order to avoid misunderstandings. For example, one could say: "Mareike A is a better craftsman than Sven, while Mareike B is worse than Sven."."
Evaluation:
In my (spontaneous) opinion this is a perfect answer that needs no addition.

I have presented a more comprehensive dialogue with a worse ChatGPT version than the current one in an earlier contribution, which also mentioned programs like DEEPL. With the help of programs such as DALL-E or Midjourney, new and arbitrary images can still be created in unprecedented quality. I have also reported on this image-forming artificial intelligence. The image above in this contribution was also generated in this way.

Protect your own content

Already a few months ago someone from Germany described that he had taken measures to protect his website's contents against being sucked up by an AI.

I see essentially the following possibilities to protect my own content against KIs like ChatGPT. What do I gain if others use my performance without getting anything in return?

The possibilities for protection against crawlers are particularly:

• Payment barrier: Will also come for higher-quality content of smaller websites sooner or later; Large news sites are already using this model.
• Anti-Bot-Protection