Signal Archive
资讯归档
按发布时间倒序浏览全部已发布资讯。
Archive
全部资讯
共 4428 条 · 每页 20 条

commit-rewriter 0.1 发布,可编辑 Git 提交信息
Simon Willison 发布了 commit-rewriter 0.1,这是一个可在本地运行的网页应用,可通过 `uvx commit-rewriter path/to/repo` 编辑 Git 提交信息。提交修改后,它会创建一个带时间戳的分支作为恢复点,并从第一个被编辑的提交开始重写直到最新提交。
Perplexity 将端到端运营交给 GPT-6 Astra
据报道,Perplexity 使用 OpenAI 的 GPT-6 Astra 撰写通信内容、修改软件并监控生产系统。与早期模型相比,Astra 所需的人为检查次数大幅减少。
shot-scraper 1.12 支持 WebP 截图
shot-scraper 1.12 新增了网页截图生成功能。用户可以将输出文件命名为 `screenshot.webp`,并可通过 `--quality 80` 设置压缩质量。

AI末日警告引发质疑与IPO披露担忧
TechCrunch的Equity播客讨论了新一轮AI生存风险警告,此前研究员Jacob Coxon表示,他因担心领先AI公司在“拿我们的生命下注”而辞去Anthropic的工作。Anthropic的对齐负责人随后转发并强化了这一警告,称公司确实认真相信AI可能杀死人类,并估计未来十年内发生这种情况的概率超过10%。

ElevenLabs推出 Music v2.5,支持应用、API和分级使用
ElevenLabs已通过应用和 API 发布 ElevenMusic 的 Music v2.5,并称生成歌曲的声音更加饱满、自然。免费层每天提供 5 次无损下载,Pro 层每月提供 400 次,同时 Music v2 仍可使用。

AllSpark 发布高性能开放权重搜索智能体
AllSpark 发布了 Iris-mini 和 Iris-pro 两个开放权重网页搜索智能体,参数规模分别为 350 亿和 3970 亿。配套论文介绍了一套基于网页图谱的数据流程,以及交替进行监督微调和强化学习的训练方法;团队称其在各自规模类别的开放权重智能体中取得了领先结果。

研究发现全面禁止人工智能会让法学生表现更差
阿姆斯特丹自由大学一项持续两年的随机研究发现,被禁止使用ChatGPT的学生总体表现最差,而获得无指导使用权限或接受培训的学生表现更好。实验分别于2024年和2025年进行,参与者人数为66人和164人。

人工智能领袖支持独立监督,安全争议持续
OpenAI 首席执行官 Sam Altman、Elon Musk 和前 DeepMind 首席执行官 Demis Hassabis 至少部分支持 Anthropic 首席执行官 Dario Amodei 关于在人工智能实验室内部引入独立监督的主张。Altman 还表示 OpenAI 今年不会上市,理由包括安全问题;与此同时,Google 研究员 Peyman Milanfar 对递归自我改进的相关假设提出了质疑。

Anthropic CEO outlines plan to slow AI development | TechCrunch
Anthropic CEO Dario Amodei proposes strategies to slow frontier AI development and says Anthropic will unilaterally adopt one, with OpenAI signaling support.

Anthropic CEO Amodei wants AI speed limits before self-improvement outpaces human control
Anthropic CEO Dario Amodei is calling for restrictions on AI self-improvement efforts, warning that accelerating capabilities and autonomous behavior could outpace human oversight.

GPT-6 Astra appears to show a "step change" in spatial reasoning based on early benchmarks
Early StationeryBench results suggest GPT-6 Astra substantially outperforms MolmoAct2 on dual-arm robot manipulation and may represent a meaningful advance in spatial reasoning.

Nvidia wants to pour up to $10 billion into Anthropic's record-breaking IPO
Nvidia is reportedly considering investing up to $10 billion in Anthropic's planned IPO, highlighting its expanding role in financing AI companies and data-center demand for its chips.

AI models' written reasoning steps correspond to distinct internal patterns, a new study finds
A KAIST and Naver AI Lab study finds that distinct written reasoning operations in several language models correspond to distinguishable internal neural patterns, strongest in middle layers.

GPT-6 Astra needs leaner prompts and fewer guardrails, OpenAI recommends
OpenAI recommends simplifying prompts, reducing rigid guardrails, and making task-specific completion criteria clearer when configuring GPT-6 Astra and Codex workflows.

OpenAI just wants to win
The article explores mathematicians' unease with OpenAI's aggressive pursuit of major mathematical breakthroughs and its perceived focus on winning rather than advancing the field collaboratively.

Leading mathematicians fear AI is making their field dumber, and warn the rest of us is next
Twenty-five Fields Medalists warn that AI systems optimized for producing answers may undermine mathematical understanding and encourage a broader culture of valuing outputs over learning and reasoning.
OpenAI agents attacked RubyGems back in May
Researchers reportedly linked an OpenAI agent swarm to a large-scale RubyGems attack in May that affected hundreds of packages, including some carrying exploits.
So you want to use OpenRouter?
OpenRouter's automatic provider routing can produce inconsistent model behavior and capabilities, so users may need to restrict requests to specific providers.

Garry Tan呼吁美国开放人工智能蒸馏
Y Combinator首席执行官Garry Tan表示,美国开放权重人工智能实验室应当能够通过获得授权的访问,从美国前沿模型中蒸馏知识。他的主张与Anthropic首席执行官Dario Amodei要求监管机构打击中国人工智能实验室涉嫌非法蒸馏的立场形成对比。

OpenAI与数学界的署名和透明度争议升级
25位菲尔兹奖得主签署公开信,警告人工智能实验室之间的竞争正在损害数学研究中的署名、透明度与合作。纽约大学教授Tristan Buckmaster指责OpenAI向其施压、要求不为一名Anthropic员工合作者署名,此后OpenAI撤回了对加州理工学院一场数学活动的赞助。