November 20, 2025 — BGR published a user‑focused review naming seven AI chatbots that users rated superior to ChatGPT on certain tasks. The article grouped the "winners" into four categories: two Gemini models from Google, two models labeled DeepSeek, two Grok models from xAI, and Magistral from French startup Mistral.
The article noted that although OpenAI's ChatGPT, thanks to its revolutionary launch and massive user base (accounting for nearly half of global AI chatbot usage), still dominates the global market, a recent independent study centered on user experience shows that the industry giant has been outperformed by several competitors in actual human–AI interaction.
A user survey conducted by UK company Prolific, based on its "Humaine" benchmark, found that after evaluating 28 mainstream AI models, ChatGPT‑4.1 ranked only eighth. The ranking shows that while ChatGPT's reach is unrivaled, new competitors are rapidly rising and winning user favor in perceived interaction quality.
The top‑ranked AI models included:
- Gemini 2.5 Pro (Google)
- DeepSeek v3 (DeepSeek, China)
- Magistral Medium (Mistral AI, France)
- Grok 4 (xAI)
- Grok 3 (xAI)
- Gemini 2.5 Flash (Google)
- DeepSeek R1 (DeepSeek)
- ChatGPT‑4.1 (OpenAI)
- Gemma (Google)
- Gemini 2.0 Flash (Google)
Prolific said its "Humaine" benchmark was created to fill gaps in traditional AI performance testing (which often focuses on pure technical metrics) by shifting the evaluation to the values users actually care about. The study invited nearly 25,000 diverse participants from the US and UK to perform blind tests, comparing pairs of different AI models across multi‑turn dialogues to ensure results were based on real user experience.
The evaluation covered four key dimensions:
- Core task performance and reasoning: answer accuracy and understanding of user intent.
- Interaction fluency and adaptability: ability to handle multi‑turn conversations and maintain contextual coherence.
- Communication style and presentation: clarity of information and the chatbot's level of "personality".
- Trust, ethics and safety: users' perceptions of transparency and safety of responses.
The results strongly indicated that Google's Gemini 2.5 Pro is users' top‑choice AI chatbot. It performed well across multiple metrics and user groups, particularly in more subjective areas such as tone, clarity and conversational adaptability, demonstrating a strong "human touch" and leadership in human‑centered AI. Meanwhile, DeepSeek v3 from China stood out, ranking second overall and taking first place in the "communication style and presentation" category.
The study sketches a new competitive landscape for AI chatbots. Although ChatGPT still holds an absolute advantage in user numbers and total usage — and may excel in other technical tests (such as passing medical/scientific entrance exams) — the Humaine rankings clearly reflect that when users evaluate anonymously without brand bias, other vendors including Google, DeepSeek and Mistral are making significant strides in delivering more humanlike, smoother and more trustworthy conversational experiences. For users seeking superior interaction quality, this feedback‑based list provides an important reference.