First lawsuit against a search engine based on Artificial Intelligence

Second lawsuit against AI

Yesterday, Pablinux warned us about the need to be careful with the responses from Bard, Google's Artificial Intelligence tool. It seems things aren't much better for Microsoft, with its Bing search engine and ChatGPT, as they are the subject of the first lawsuit against an AI-based search engine.

Search engines have already been sued in several countries for the results they displayed. In some cases, the courts acquitted them, and in others, they were held responsible for indexing material published by third parties. But now we face an unprecedented situation.

The first lawsuit against a search engine based on Artificial Intelligence

It all started when someone thought to search for Jeffery Battle on Bing, Microsoft's search engine that uses the technology behind ChatGPT. The thing is, there are two important enough to appear in search results.

One is a respectable business leader and university professor retired with honors from the United States Air Force, the second, who also has the second name of Leon, is in prison for wanting to join Islamic terrorists.

For Bing, however, it is the good professor who is imprisoned for wanting to join the Taliban . And the "however" is literal. It is the transition chosen by the AI ​​to merge the two biographies, unequivocally implying that they are the same person.

Professor Battle filed the defamation lawsuit and is so confident of success that he decided to forgo legal representation, something permitted under the country's law. However, analysts are not so convinced of his chances.

According to US law:

No provider or user of an interactive computer service shall be treated as a publisher or speaker of any information provided by another information content provider."

However, the plaintiff argues that not only did he inform Microsoft of the problem, but by processing the information he loses the immunity from citation to become the issuer.

This is the second lawsuit against an Artificial Intelligence tool. The first, directed against OpenAI is for providing false information about a lawsuit.

The Defendant Who No One Accused

Fred Riehl is a journalist who decided to subscribe to ChatGPT to make his work easier. One day he asked the tool to give him a summary of a lawsuit and provided the link where he could find it—the plaintiff's official website.

The response was that the lawsuit accused a former finance director (whom he named) of embezzlement . The problem is that this person never held that position, and the lawsuit has nothing to do with embezzlement.

Riehl, who I assume knew what the lawsuit was about, asked him to quote the relevant paragraph in which Walters was accused, which, to his surprise, ChatGPT did.

When the journalist asked him to show the full complaint, ChatGPT produced a text that had nothing to do with the original; not even the case number matched.

However, according to experts, Mr. Walters' chances of winning the lawsuit are doubtful. The crime of defamation requires that:

  • The plaintiff can show that the defendant acted knowing that the statement was false, or knew that the statement might be false, but negligently failed to act on this knowledge.
  • The plaintiff can prove that he was harmed by the statement. Examples of this would be job loss, loss of family or business relationships.

On the ChatGPT website, at least in its free version, it warns that the results may have errors. If OpenAI is aware of the magnitude that these errors can have, it should be tested. What is certain is that at the moment I am not going to change my lawyer for an AI, ChatGPT does not seem able to avoid being sued.


Add as preferred source in Google