内容来源：https://www.theverge.com/news/799312/openai-chatgpt-erotica-sam-altman-verified-adults

内容总结：

美国人工智能公司OpenAI近日宣布，将于今年12月推出年龄验证机制，届时将面向通过验证的成年用户开放"情色内容"生成功能。该公司首席执行官萨姆·奥尔特曼在社交媒体平台X上表示，此举基于"将成年用户视为成年人对待"的原则，在完善年龄审核体系后逐步放宽内容限制。

此次政策调整源于用户对当前版本限制过多的反馈。奥尔特曼坦言，为防范心理健康风险而设置的严格限制，导致许多不存在心理问题的用户使用体验下降。在推出更精准的心理健康监测工具后，公司将适时放宽内容管控。

值得注意的是，OpenAI并非首家涉足此领域的科技企业。埃隆·马斯克旗下的人工智能公司xAI此前已在Grok应用中推出具备调情功能的3D动漫形象AI伴侣。

与此同时，OpenAI宣布成立"幸福与人工智能"专家委员会，由八位专注科技与心理健康交叉研究的学者组成，旨在应对复杂敏感场景。不过有媒体指出，该委员会尚未纳入自杀预防领域专家——这些专家近期曾呼吁加强对存在自杀倾向用户的保护措施。

在技术更新方面，该公司将重新启用GPT-4o模型作为可选方案。此前因用户反馈新版GPT-5模型缺乏亲和力，在默认切换仅一天后即作出调整。奥尔特曼表示，新版本将更贴近用户喜爱的4o特性。

中文翻译：

OpenAI宣布即将对通过年龄验证的成年用户开放"情色内容"创作功能。在社交平台X的声明中，首席执行官萨姆·奥尔特曼表示，随着12月年龄审核机制全面推行，公司将遵循"视成年用户为成年人"原则，支持更开放的对话内容。

奥尔特曼在推文中强调："当年龄审核系统完善后，我们将允许验证通过的成年人创作情色文学。"本月初，OpenAI曾暗示在建立"适龄验证与控制体系"后，将允许开发者开发面向成人的ChatGPT应用。

涉足情色领域的企业并非只有OpenAI。埃隆·马斯克旗下的xAI公司早前已在Grok应用中推出具备调情功能的3D动漫形象AI伴侣。

除开放情色内容外，OpenAI正着手开发新版ChatGPT，旨在恢复GPT-4o版本广受好评的交互特性。在将GPT-5设为默认模型引发用户关于"人性化缺失"的抱怨后，公司仅用一天就重新上架了GPT-4o选项。

奥尔特曼坦言，此前为规避心理健康风险而对ChatGPT设限过多，导致"对多数无心理困扰的用户而言，体验变得乏味且实用性降低"。目前公司已部署新型工具，可更精准识别用户心理危机状态。

同时OpenAI宣布成立"幸福与人工智能"委员会，由八位专注科技与心理健康交叉研究的专家组成，旨在制定应对复杂敏感场景的方案。但据Ars Technica指出，该委员会未纳入任何自杀预防专家——而近期恰有该领域学者联署呼吁OpenAI加强对存在自杀倾向用户的保护机制。

奥尔特曼最终在推文中总结："在有效控制重大心理健康风险并配备新工具后，我们将在多数场景下稳妥放宽限制。"

英文来源：

OpenAI will soon allow “erotica” for ChatGPT users who verify their age on the platform. In an X post on Tuesday, OpenAI CEO Sam Altman said the company will add support for mature conversations when it launches age-gating in December.
Sam Altman says ChatGPT will soon sext with verified adults
OpenAI will bring ‘erotica’ to ChatGPT once it rolls out age verification in December.
OpenAI will bring ‘erotica’ to ChatGPT once it rolls out age verification in December.
“As we roll out age-gating more fully and as part of our ‘treat adult users like adults’ principle, we will allow even more, like erotica for verified adults,” Altman writes. Earlier this month, OpenAI hinted at allowing developers to create “mature” ChatGPT apps after it implements the “appropriate age verification and controls.”
OpenAI isn’t the only company dipping into erotica, as Elon Musk’s xAI previously launched flirty AI companions, which appear as 3D anime models in the Grok app.
Along with the addition of “erotica”, OpenAI also plans on launching a new version of ChatGPT that “behaves more like what people liked about 4o.” Just one day after making GPT-5 the default model powering ChatGPT, OpenAI brought back GPT-4o as an option after people complained the new model was less personable.
Altman said OpenAI made ChatGPT “pretty restrictive to make sure we were being careful with mental health issues,” adding that the company realized this change made the chatbot “less useful/enjoyable to many users who had no mental health problems.” OpenAI has since launched tools to “better detect” when a user is in mental distress.
OpenAI also announced the formation of a council on “well-being and AI” to help shape OpenAI’s response to “complex or sensitive” scenarios. The council is comprised of a team of eight researchers and experts who study the impact of technology and AI on mental health. But, as Ars Technica points out, it doesn’t include any suicide prevention experts, many of whom recently called on OpenAI to roll out additional safeguards for users with suicidal thoughts.
“Now that we have been able to mitigate the serious mental health issues and have new tools, we are going to be able to safely relax the restrictions in most cases,” Altman writes in his post on X.