我们仍然不知道人们究竟如何使用 AI

此外，被研究人员标记为敏感使用的交流——即包含可能有害或受限内容的交流，包括性骚扰和仇恨言论——有所下降。这或许表明，平台整体上部署了更有效的安全防护措施。

AI Observatory 还发现，不同模型的 AI 使用方式存在显著差异。根据所使用的工具不同，用户在讨论主题、交互风格、对话结构，以及敏感使用场景的发生概率和类型方面都不尽相同。

例如，研究人员发现，人们更常使用 Grok 和 Gemini 进行信息检索。Grok 尤其常被用于获取新闻和政治信息，但错误信息也往往集中于该平台。（这与其他研究一致；这些研究同样显示，错误信息如何轻易地在 Grok 上扩散。xAI 未回应置评请求。）

与此同时，人们更倾向于使用 Anthropic 的产品进行编程，使用 Gemini 进行社交和角色扮演用途，以及使用 ChatGPT 获得作业帮助。

即便是同一模型的不同版本之间，也存在差异。研究人员发现，当 ChatGPT 由 GPT-3.5 驱动时，人们的对话更短；而由 GPT-4o 驱动时，对话则更长、迭代性更强——考虑到该版本后来因导致情感成瘾而广为人知，这一点并不令人意外。****

不过，公司的报告往往无法捕捉不同模型之间、甚至同一模型内部的这些细微差别。“没有任何一份单一公司的报告能够讲述完整故事，”与 Reuel 共同领导这项研究的 MIT Media Lab 近期博士毕业生 Shayne Longpre 表示。

为创建 AI Observatory，Reuel 与来自 MIT、斯坦福、Data Provenance Initiative及其他机构的研究人员，汇总了先前研究收集的 7 个真实世界数据集中的 24,521 段对话、共 85,633 个对话轮次（即用户提示词及相应的 AI 回复）。这些对话来自 2023 年至 2025 年间与包括 ChatGPT、Gemini、Claude 和 Grok 在内的 52 种不同模型互动的 5,000 名用户。

但与大型实验室自身能够获取的数据相比，这些对话只是沧海一粟。例如，最新的 Anthropic Economic AI Index基于对 100 万段 Claude 对话的分析；OpenAI 关于人们如何使用 ChatGPT的报告则分析了 150 万段对话。

Additionally, exchanges that the researchers labeled as sensitive use—meaning ones with potentially harmful or restricted content, including sexual harassment and hate speech—dropped. That might suggest that platforms were generally deploying more effective safeguards.

The AI Observatory also found that AI use looked significantly different depending on the model. Depending on the tool, users ranged in topics, interaction styles, conversation structures, as well as both the likelihood and type of sensitive use cases.

For example, the researchers found that people used Grok and Gemini more frequently for information retrieval. Grok, in particular, was especially popular for information on news and politics, but it was also where misinformation tended to concentrate. (This is consistent with other research that has also shown how readily misinformation proliferates on Grok. xAI did not respond to a request for comment.)

Meanwhile, people were more likely to turn to Anthropic for coding, Gemini for social and roleplay uses, and ChatGPT for homework assistance.

There were even differences among different versions of the same model. Researchers found that people had shorter conversations with ChatGPT when it was powered by GPT-3.5, and longer and more iterative ones with GPT-4o—which makes sense given that that version became known for leading to emotional addiction.****

Company reports, however, didn’t tend to capture these nuances across or even within their own models. “No single company report tells the whole story,” says Shayne Longpre, a recent PhD graduate from the MIT Media Lab who co-led the research with Reuel.

To create the AI Observatory, Reuel and researchers from MIT, Stanford, the Data Provenance Initiative, and other institutions, aggregated 24,521 conservations across 85,633 conversational turns (that is, the user prompt and corresponding AI response) from seven real-world datasets collected by previous research. These conversations came from 5,000 users interacting with 52 different models, including ChatGPT, Gemini, Claude, and Grok, between 2023 and 2025.

But these conversations are a drop in the proverbial bucket compared to the data that the big labs themselves have access to. The latest Anthropic Economic AI Index, for example, is based on analysis of 1 million Claude conversations; OpenAI’s report on how people are using ChatGPT analyzed 1.5 million conversations.