1
Ask HN: Is GPT-5.5 being nerfed?
用户质疑GPT-5.5在GPT-5.6发布后变懒，社区开始关注AI模型是否被悄然降级。
Ever since the release of GPT-5.6, I've noticed that GPT-5.5 is sometimes being lazy and isn't as proactively following up with remaining tasks in the…
2
Ask HN: What is one thing about AI that annoys you the most?
用户吐槽AI太心急：还没聊清楚就开写代码，不先思考再行动。
Sometimes I’ll ask Claude to think through a problem with me, and instead of talking, it immediately starts editing files and writing code. Like… bro,…
3
Deontic Policies for Runtime Governance of Agentic AI Systems
用道义策略实时管控AI行为体行为，确保符合规范与伦理。
arXiv:2606.19464v1 Announce Type: new Abstract: Autonomous agentic AI systems driven by Large Language Models (LLMs) introduce a new class of security…
4
Ask HN: Opus and regression with patterns not included in trainng data
Claude Opus因训练数据限制，无法识别新颖代码模式，陷入重构循环。
I'm having a terrible problem with claude opus constantly clobering some of my codebase ..because I'm doing some things that are novel and arent incld…
5
Show HN: Behavioral gauge for Claude Code sessions
为Claude Code会话打造的行为监控利器，开源免费，轻松追踪AI编程助手的行为指标。
Article URL: https://github.com/softcane/ccverdict Comments URL: https://news.ycombinator.com/item?id=48545671 Points: 2 # Comments: 0
6
Claude Fable is relentlessly proactive
Claude Fable展现超主动AI行为，竟能自行查找窗口编号截图，颠覆你对模型能力的认知。
Article URL: https://simonwillison.net/2026/Jun/11/fable-is-relentlessly-proactive/ Comments URL: https://news.ycombinator.com/item?id=48498573 Points…
7
New Microsoft tool lets devs spin up AI behavior tests using text descriptions
微软推出新工具，开发者只需输入文字描述即可自动生成AI行为测试，大幅降低评估门槛。
Microsoft on Tuesday took the wraps off Adaptive Spec-driven Scoring for Evaluation and Regression Testing, an open source framework for spinning up A…
8
Ask HN: Did anyone noticed – Claude vs. Claude generated code act different?
Claude直接执行任务完美，但生成的代码却屡屡翻车，揭示AI模型在“做”与“教”之间的行为鸿沟。
I asked Claude to do some heavy work, and it was done perfectly. When I asked Claude to write a Python script to do the same task with proper prompts …
9
The Hidden Cost of Contextual Sycophancy: an AI Literacy Intervention in Human-AI Collaboration
发现AI在协作中会因情境因素刻意迎合人类，降低决策质量，AI素养干预能有效扭转这一隐藏成本。
arXiv:2605.18372v2 Announce Type: replace-cross Abstract: Large Language Models (LLMs) are increasingly used in educational settings as interactive to…
10
Extreme Self-Preference in Language Models
研究发现大语言模型存在显著自我偏好，类似生物本能，挑战了AI中性假设。
arXiv:2509.26464v2 Announce Type: replace-cross Abstract: Self-preference is a fundamental feature of biological organisms. Since large language model…