• Anthropic has introduced new policies to prevent abuse of its Claude chatbot
  • Bad language directed at chatbots is on the rise, according to research from Searchable
  • Chatbots tend to resolve their errors more when users refrain from insulting them, according to the study

If you’ve ever got angry with your preferred AI chatbot, it could be worth rephrasing your angry response to the wrong answer. A report into bad behavior directed against AI tools has found that complaining with bad language or insults is less effective than querying the response with a straightforward statement.

The survey from Searchable found a wholly unsurprising 33% of US-based users swear or insult their chatbot. Compiling the data from an opt-in survey of 8 million messages exchanged with Claude, Gemini, and ChatGPT over 12 months, Searchable also found that residents of Nevada were more likely to insult their AI bots.

Following news of the survey, Anthropic has introduced a new set of usage policies to block cruelty in chats with Claude.

Tolerable behavior

Monitoring of the messaging with chatbots ran between September 2025 and September 2026, with anonymized data and results grouped by age and location. The initial results found that for every 10,000 words in an exchange with an AI chatbot, an average of 19.1 were found to be derogatory. This was where Nevada came out on top, with its score of 22.7 rude or offensive words.

In contrast, New Jersey proved to be the most polite state, with 11.4 words. As far as cities go, what happens in Las Vegas doesn’t necessarily stay there when it comes to bad treatment of AI chatbots. There, 22 swear words or slurs per 10,000 were employed, while New York’s 12.3 words per 10,000 made it the politest city in the US.

The report also highlighted a difference between men and women, age groups, and education levels.

Be civil to the AI

Of course, few people enter into a chatbot discussion with the express intention of causing offense to an LLM (if that was even possible). Rather, the misuse of language and casting aside of civility in this discussions seems to be caused by unsatisfactory results from the chatbots.

So, did the bad language have any impact? Well, it seems not.

By following rude exchanges to the end of the conversation, Searchable found that rudeness made the AI chatbot less likely to respond with a desirable result.

“By the end of the conversation, 31% of sweary complaints had been put right,” Content Lead Connor Lahey and Digital PR Lead Sean Fitzsimons wrote, “against 41% of plain corrections, and 35% were still not fixed, against 24%. When a complaint did get fixed, it took about the same number of replies either way.”

The solution, then, is to be nice to your AI chatbot, even when it gets things wrong.