Technology news and discussion on TalkFuse.
A third of US AI chatters are hot at bots, hurling insults at LLMs
A survey found that 33 per cent of US chatbot users swear at or insult AI assistants, prompting Anthropic to introduce a rule against sustained, needless cruelty towards its Claude model. The policy took effect on 12 November and reflects the company’s broader interest in how people treat AI systems, while distinguishing gratuitous abuse from interactions with a discernible purpose.
Searchable analysed eight million messages sent to ChatGPT, Gemini and Claude by an opted-in, anonymised US panel between September 2025 and September 2026. Chatbots resolved users’ initial complaints in 24 per cent of cases when insulted, compared with 35 per cent when users did not insult them; 52 per cent of abusive messages followed problems such as errors or misunderstood instructions, and 8 per cent came after refusals or safety warnings. Anthropic says its rule targets repeated cruelty without a clear purpose, while attempts to bypass safeguards remain prohibited under separate platform abuse rules.
In short
- One in three US chatbot users insulted or swore at AI assistants.
- Insulted chatbots resolved complaints less often in the survey.
- Anthropic’s new rule targets sustained, needless cruelty towards Claude.
Steelman: both sides, in good faith
The strongest fair case each way — we don't pick a winner.
The case for
A company can reasonably set limits on sustained, purposeless abuse to shape the kind of interactions it supports and discourage habits it regards as harmful. The survey also suggests insults are associated with worse complaint resolution, so a boundary could encourage clearer, more effective exchanges while still allowing criticism tied to a specific problem.
The case against
Users can reasonably expect to express frustration when a chatbot makes errors or misunderstands instructions, and the survey indicates that many insults follow such problems. Restricting how users speak to a system could feel paternalistic or blur the line between gratuitous cruelty and forceful criticism, especially when the system itself has failed to help.
Explained simply
Large language models (LLMs) are AI systems that respond to written prompts, such as ChatGPT, Gemini and Anthropic’s Claude. People use them for tasks ranging from answering questions to drafting text, but they can make mistakes or misunderstand instructions.
A US survey found that some chatbot users swear at or insult these systems, often after a problem with the response. Anthropic makes Claude and has taken an interest in how people interact with AI; its approach distinguishes sustained insults with no clear purpose from criticism tied to a specific problem.
The issue matters because the way people communicate with chatbots may affect how useful the exchange is: the survey found complaints were resolved less often when users insulted the AI. Rules against abuse also raise questions about how companies should set boundaries for interactions with systems that are not people.
AI Americas Health Medicine Technology World