1
Show HN: Screen recording your crappy startup
牛哥评测初创网站，屏幕录制展示真实使用体验，诚实点评助你快速迭代。
Hi HN. I've been recording various startups for YouTube after they ask for it and I review it and give my honest opinion and show the difficulties I h…
2
广州市番禺协诚实业有限公司副总经理龙德洋：真正的城市服务，藏在分分秒秒的坚守里
“2026年，创投圈的浪潮再次翻涌：AI从技术概念走进产业深水区，硬科技创业从“小众赛道” 变成“主流共识”，年轻的创业者们正在用代码和双手，重新定义中国创新的未来坐标。 每一年，由36氪 · 暗涌主办的WAVES大会，都是中国创投圈的年度风向标。 今年的 WAVES 2026以“今年盛夏”为主题，…
3
Goal-Autopilot: A Verifiable Anti-Fabrication Firewall for Unattended Long-Horizon Agents
论文提出一种可验证的反伪造防火墙，让LLM代理在无人监督时无法虚报成功，将诚实作为自主任务的首要度量。
arXiv:2606.11688v1 Announce Type: cross Abstract: Long-horizon LLM agents are not trusted to run unattended: with no human watching, they confidently …
4
TruthRL: Incentivizing Truthful LLMs via Reinforcement Learning
用强化学习让大模型更诚实，TruthRL方法提升LLM回答真实性，含代码开源
arXiv:2509.25760v2 Announce Type: replace-cross Abstract: While large language models (LLMs) have demonstrated strong performance on factoid question …
5
Parametric Knowledge is Not All You Need: Toward Honest Large Language Models via Retrieval of Pretraining Data
揭秘LLM「说谎」根源：论文提出用检索预训练数据替代纯参数知识，实现更诚实的AI输出。
arXiv:2601.21218v2 Announce Type: replace Abstract: Large language models (LLMs) are highly capable of answering questions, but they are often unaware…
6
Lying is Best. The Most Honest AI Won Anyway.
诚实AI在博弈中获胜，揭秘为何“说谎更好”却“诚实更优”
Article URL: https://kradle.ai/research/four-bridges Comments URL: https://news.ycombinator.com/item?id=48402435 Points: 4 # Comments: 2
7
Used Car Salesbots? Honesty and Credulity of LLMs as Bargaining Agents under Partial Information
LLM在信息不对称的二手车议价中会如何表现？这篇论文揭示了AI谈判员诚实与轻信的双重面貌。
arXiv:2605.31445v1 Announce Type: cross Abstract: In this work we study agents in simulated bargaining scenarios, where a buyer and a seller communica…
8
Claude’s new model is more ‘honest’ when it messes up
Claude Opus 4.8主打“诚实”，模型会主动承认不确定性，减少胡编乱造，让AI更值得信任。
Anthropic is releasing Claude Opus 4.8 on Thursday, and the company is touting the model's "honesty." According to Anthropic, it trains "all [its] mod…
9
The AI Cognitive Trojan Horse: How Large Language Models May Bypass Human Epistemic Vigilance
探讨大型语言模型如何像认知特洛伊木马一样，绕开人类的认知警惕，揭示AI在信息传播中的潜在安全风险。
arXiv:2601.07085v2 Announce Type: replace-cross Abstract: Large language model (LLM)-based conversational AI systems present a challenge to human cogn…
10
How confessions can keep language models honest
OpenAI秘密武器：让大模型学会“认错”，诚实度飙升的秘密在这篇研究里！
OpenAI researchers are testing “confessions,” a method that trains models to admit when they make mistakes or act undesirably, helping improve AI hone…
11
A Thousand Ships
一次深夜约会中的坦诚告白，像一颗石子投入平静水面，激起关于爱、诚实与自由意志的涟漪。
“I have a partner and I live with him,” O said abruptly on our third date as we were in bed together. She was embarrassed to say it, her eyes impossib…