公开采集：原帖显示38条评论，本次可读取10条。下方来源为同一帖子的另一原生标识，用来归组讨论，并非另一篇帖子。以下为编辑摘要而非逐字引述；未展开评论及未确认的回复父子关系不作推测。
1. Matthew Kilkenny · 2026-10-29T17:35:59.102Z
链接一篇去年的帖子，追问情况变好还是变坏。
https://www.linkedin.com/feed/update/urn:li:activity:7521621606486470656/?commentUrn=urn%3Ali%3Acomment%3A%28urn%3Ali%3AugcPost%3A7521621353028902912%2C7521625897863561216%29
2. Matthew Kilkenny · 2026-10-29T18:32:06.723Z
追问对模拟爱情产生情感依赖时，谁承担道德责任。
https://www.linkedin.com/feed/update/urn:li:activity:7521621606486470656/?commentUrn=urn%3Ali%3Acomment%3A%28urn%3Ali%3AugcPost%3A7521621353028902912%2C7521640022685667328%29
3. Matthew Kilkenny · 2026-10-29T18:37:08.016Z
追问 Dessa 是研究警示，还是走向广泛部署的预告。
https://www.linkedin.com/feed/update/urn:li:activity:7521621606486470656/?commentUrn=urn%3Ali%3Acomment%3A%28urn%3Ali%3AugcPost%3A7521621353028902912%2C7521641286400057344%29
4. Matthew Kilkenny · 2026-10-29T18:40:48.888Z
追问是否允许儿童与 AI 伴侣建立浪漫依恋。
https://www.linkedin.com/feed/update/urn:li:activity:7521621606486470656/?commentUrn=urn%3Ali%3Acomment%3A%28urn%3Ali%3AugcPost%3A7521621353028902912%2C7521642212804415489%29
5. Matthew Kilkenny · 2026-10-29T18:43:32.647Z
追问如何帮助儿童区分人类情感与算法模拟。
https://www.linkedin.com/feed/update/urn:li:activity:7521621606486470656/?commentUrn=urn%3Ali%3Acomment%3A%28urn%3Ali%3AugcPost%3A7521621353028902912%2C7521642899659444225%29
6. Matthew Kilkenny · 2026-10-30T16:06:03.226Z
认可 Henry 的能动性观点，同时追问谁控制亲密关系架构与心理脆弱信息。
https://www.linkedin.com/feed/update/urn:li:activity:7521621606486470656/?commentUrn=urn%3Ali%3Acomment%3A%28urn%3Ali%3AugcPost%3A7521621353028902912%2C7521965653780500480%29
7. piotr G. · 2026-10-29T17:52:53.810Z
担忧商业操纵及社交媒体历史教训；属于评论者观点。
https://www.linkedin.com/feed/update/urn:li:activity:7521621606486470656/?commentUrn=urn%3Ali%3Acomment%3A%28urn%3Ali%3AugcPost%3A7521621353028902912%2C7521630153853292544%29
8. Erika Warfield · 2026-10-29T20:34:52.790Z
既感到惊叹又不安，担心儿童将反应误认为真实感情。
https://www.linkedin.com/feed/update/urn:li:activity:7521621606486470656/?commentUrn=urn%3Ali%3Acomment%3A%28urn%3Ali%3AugcPost%3A7521621353028902912%2C7521670918209699841%29
9. Henry Wang · 2026-10-30T11:15:11.972Z
Henry 区分逼真声音与可靠性，追问谁控制记忆披露，讨论监管主张并呼吁独立评估，同时链接自己的长帖。
https://www.linkedin.com/feed/update/urn:li:activity:7521621606486470656/?commentUrn=urn%3Ali%3Acomment%3A%28urn%3Ali%3AugcPost%3A7521621353028902912%2C7521892457916104704%29
10. Christopher Wright · 2026-10-29T19:24:26.009Z
对盈利动机与 AI 相关心理伤害提出担忧；该评论不是临床评估。
https://www.linkedin.com/feed/update/urn:li:activity:7521621606486470656/?commentUrn=urn%3Ali%3Acomment%3A%28urn%3Ali%3AugcPost%3A7521621353028902912%2C7521653189805502464%29
