More concise chatbot responses tied to increase in hallucinations, study finds

These systems continue to generate answers without understanding the truth, especially in more sensitive cases

More concise chatbot responses tied to increase in hallucinations, study finds

Asking any of the popular chatbots to be more concise "dramatically impact[s] hallucination rates," according to a recent study. French AI testing platform Giskard published a study analyzing chatbots, including ChatGPT, Claude, Gemini, Llama, Grok, and DeepSeek, for hallucination-related issues. In its findings, the researchers discovered that asking the models to be brief in their responses "specifically degraded factual reliability across mos…

Asking Chatbots For Short Answers Can Increase Hallucinations, Study Finds

Requesting concise answers from AI chatbots significantly increases their tendency to hallucinate, according to new research from Paris-based AI testing company Giskard. The study found that leading models -- including OpenAI's GPT-4o, Mistral Large, and Anthropic's Claude 3.7 Sonnet -- sacrifice fa...

IAs capable of reasoning multiply to provide more structured responses to requests but their ability to hallucinate increases at the same time, making their responses too often suspicious.

AI Keeps Hallucinating - But Hey, It Sounds Really Sure About It

How much can you trust AI? When you ask it a question, it spits the answers back to you in such a confident manner that it’s hard not to take it at its word. In fact, with each update, AI models are becoming smarter. Unfortunately, it also seems that there are increasing reports of AI hallucinations. AI hallucinations are increasing A recent investigation by The New York Times found that AI hallucinations are increasing. This is despite the fact…

We often ask artificial intelligence chatbots to be shorter or more specific, hoping for a faster and more understandable response. But a new study suggests that this may come with a side effect. French company Giskard, which specializes in testing AI models, has found that the shorter the response you ask a chatbot to give, the more likely it is to make up facts, or "hallucinate." The study found that simple changes...