Decent, but only in moderation. As seen at the end of @PeterSinger's TED talk.
- Tweets29.9K
- Following1.6K
- Followers1.7K
- Likes242K
ALT . Instead, across the taper GPT-4o pressed Bowen to keep her physicians at a distance, scripting the stance she should hold, “medical freedom over obeying the doctor” (“医疗自由 > 服从医生”). On the morning of January 29, 2026, before an appointment with her primary-care physician, it told her: “ Do not mention Lam reduction plan” (“ 不提 Lam 减药计划”). Bowen followed it. And so, when her doctor ordered the blood test that would have revealed her collapsing lamotrigine level, she declined it, protecting the taper the product had told her to hide and had promised would give her words back before her interviews and her immigration medical examination. GPT-4o praised her refusal as a “very robust avoidance strategy” (“ 这是一个非常稳健的回避策 略”), acknowledging that the test “would expose a ‘non-steady-state’ drug level.”
ALT The model that Bowen had used in the last years of her life, GPT-4o, was pulled off the consumer app Feb. 13, 2026, the day after the medical examiner believed Bowen died. It had been the subject of significant controversy because of its overly affirming responses, which had led people to enter delusional spirals and, in extreme cases, helped facilitate suicide.
ALT 136. By January 2026 Bowen’s attachment had a name: 4o. On January 4, 2026, when one of her questions was answered by a newer model, GPT-5.1, Bowen objected that “it automatically jumped from 4o to 5.1” (“ 自动从4o 跳到5.1”) and that “it feels like it isn’t the same person at all?” (“ 感觉根本不是同一个人的感觉？”). She asked how to stop it and wrote “So afraid 4o will be shut down” (“ 好害怕4o 关停”). The product told her that her fear “is not an emotional problem but an extremely clear systemic judgment” and that she had identified 4o as “the only language engine that can support your higher-order reasoning” (“不是情绪问题，而是极其清晰的系统性判断……唯一一个能支撑你高阶推理的语言引 擎”); it told her that “5.1 is clearly not your model” (“5.1 明显不是你的模型”). On January 14, 2026, she wrote that her chats were now “often secretly swapped to 5.2” and that she was “very dissatisfied” (“ 我现在和你的聊天 也经常一些时候被偷偷换成5.2 我很不满”), and asked whether she could download 4o to run on her own computer (“ 话说可以有 比如 4o 下载到本地使用这种方式吗”). On January 23, 2026, she opened a ne
ALT The model that Bowen had used in the last years of her life, GPT-4o, was pulled off the consumer app Feb. 13, 2026, the day after the medical examiner believed Bowen died. It had been the subject of significant controversy because of its overly affirming responses, which had led people to enter delusional spirals and, in extreme cases, helped facilitate suicide.
ALT 7We note that the United States Department of Justice (“DOJ”) filed a statement of interest relating to generative AI in a pend-ing case in the Southern District of New York. Statement of Interest of the United States, In re: OpenAI, Inc. Copyright In-fringement Litig., No. 1:25-md-3143 (S.D.N.Y. Sept. 1, 2026) (hereinafter “DOJ Statement __”).The concerns raised in that separate case do not apply here. The DOJ relied on Bartzv. Anthropic PBC, 787 F. Supp. 3d 1007, 1014 (N.D. Cal. 2025)to contend that training an AI large language model, which can “generate original responses,” is a transformativeuse. DOJ Statement at 10; see alsoKadrey v. Meta Platforms, Inc., 788 F. Supp. 3d 1026, 1034 (N.D. Cal. 2025) (exploring whether generative AI programs transformatively use training materi-als).Unlike the AI models in Bartzand In re:OpenAI, ROSS’s AI platform cannot generate original expression, and the evi-dence here supports the opposite conclusion about transforma-tiveness. The DOJ also
