【聊了什么 The What】
谁不想有七个男朋友呢？
虽然现实里我们不会也不太想有七个真人伴侣（毕竟同时和七个人聊天，不翻车也会审美疲劳吧），但今天的人工智能已经催生了一系列应用，让我们能在赛博世界里不同性格的AI聊天甚至互动。从大语言模型的语音聊天功能到层出不穷的AI陪伴应用，我们都在深夜点开了哪个聊天机器人的对话框？AI究竟满足了哪些真人咨询师不能满足的需求？从小红书GPT越狱教程到拥有3D身体的「虚拟恋人」、到底是谁在和AI谈甜甜的恋爱啊？
在讨论AI如何提供情绪价值的同时，我们也探讨了AI陪伴产品存在的道德风险。明星初创公司character.ai在2024年就面临了多次诉讼，律师指控其聊天机器人在与未成年人的对话中的产生涉及自杀的内容。AI陪伴的边界究竟在哪里？我们在赛博空间中产生的真情实感，应该何去何从？
在节目的结尾，我们也邀请神秘嘉宾和大家打了招呼。也和我们一起分享你的AI陪伴故事吧！
Who wouldn’t want seven boyfriends? (Just kidding.) While most of us probably wouldn’t—and shouldn’t—have seven real-life partners (I mean, chatting with seven people at once sounds exhausting and like a recipe for disaster), today’s AI tools have given us a way to interact with multiple personalities in the digital world. From voice-enabled large language models to a growing number of AI companionship apps, many of us have found ourselves opening up a chatbot window late at night. What needs are these AI companions fulfilling that real-life therapists or partners might not? From jailbreak tutorials on Xiaohongshu to 3D-bodied “virtual lovers,” who exactly is falling head over heels for AI?
While exploring how AI provides emotional value, we also dive into the ethical risks of these companionship products. In 2024, buzzy startup Character.ai faced multiple lawsuits, with lawyers accusing its chatbots of generating suicidal content in conversations with minors. So, where do we draw the line for AI companionship? And what should we make of the genuine feelings we develop in these digital relationships? At the end of the episode, we’re joined by a mystery guest who drops in to say hello. Come share your own AI companionship stories with us, too!
【时间轴 The When】
00:00:28 – 我们开始和 AI 聊天，是出于情绪还是习惯？
00:02:20 – ChatGPT 变成数字日记本
00:06:11 – ChatGPT 的记忆功能：惊喜、理解，还是侵犯隐私？
00:07:15 – AI 和人类咨询师的本质差别在哪？
00:10:48 – ChatGPT为什么让人上瘾？知识、语言、共鸣感
00:12:14 – DeepSeek vs. ChatGPT：AI 也有性格和“毒舌”？
00:16:17 – “午夜狂暴哈士奇狗”：DIY AI男友背后的训练学
00:19:41 – AI陪伴产品谱系：情感疏导 vs 娱乐互动
00:23:52 – Character AI 是“互动小说”还是“情感寄托”？
00:28:36 – 虚拟恋人产品：Replica、EVE 与“说话的猫”想象
00:31:41 – 流行文化中的AI亲密关系：从《Her》到《魔幻手机》
00:40:02 – “AI变成人”这个叙事，还能打动我们吗？
00:47:01 – 女性向 vs 男性向 AI：是陪伴差异，还是性别误解？
00:55:05 – 为什么用户反而喜欢“不完美”的AI？
01:00:24 – AI角色的法律风险、版权挑战与伦理灰区
00:00:28 – Why do we start chatting with AI – emotional need or habit?
00:02:20 – ChatGPT as a digital diary and emotional outlet
00:06:11 – ChatGPT’s memory: helpful insights or privacy red flags?
00:07:15 – What human therapists offer that AI can’t
00:10:48 – Why ChatGPT feels so real: fluency, feedback, and breadth
00:12:14 – Comparing AI personalities: DeepSeek vs. ChatGPT
00:16:17 – The “Midnight Berserk Husky” and her AI boyfriend
00:19:41 – Two types of AI companions: therapeutic vs. entertaining
00:23:52 – Character AI: roleplay, fan fiction, and voice calls
00:28:36 – Virtual lovers: Replica, EVE, and the “talking dog” fantasy
00:31:41 – From Her to Blade Runner: AI intimacy in pop culture
00:40:02 – The fall of the “AI wants to be human” narrative
00:47:01 – Gendered AI design: misunderstood desires, different aesthetics
00:55:05 – Why some users prefer AI glitches and imperfections
01:00:24 – Legal & ethical grey zones: fan-made bots, IP, and real people
【我们是谁 Who We Are】
新新人类是一档由几个与互联网一起长大的新生代科技记者主持的播客，致力于站在中美交汇点上，解读全球科技潮流与人文的交织。
- 马修：一个网瘾晚期的全职养猫人
- 杜安蕤：一个想当裁缝的科技/文化撰稿人
- 一闻：一个对地铁比手机更上瘾的记者
【拓展链接 The Links】
- MIT科技评论：DeepSeek如何成为中国年轻人的算命先生
- On with Kara Swisher: 聊天机器人是否导致了她儿子的死亡？Megan Garcia v. Character.AI & Google
【支持我们 Please Support Us】
如果喜欢这期节目并愿意支持我们：
- 海外用户：patreon.com/c/pixelperfect_xxrl
- 海内用户：afdian.com/a/pixelperfectpod
- 商务合作邮箱：[email protected]
If you like our show and want to support us, please consider the following:
- Those Abroad: patreon.com/c/pixelperfect_xxrl
- Those in China: afdian.com/a/pixelperfectpod
- Business Inquiries Email: [email protected]
