新信号雷达
扫描各信源刚发布的内容，并直接打开原文。这里只收待核验信号，通过证据检查后才会写入档案。
AI for Bug Bounty with VulneraMCP
Article URL: https://www.zaproxy.org/blog/2025-11-28-enhancing-zap-with-ai-for-bug-bounty-hunting/ Comments URL: https://news.ycombinator.com/item?id=48962545 Points: 1 # Comments: 0
$20/Month: The Price Ceiling Every AI Company Copied
Article URL: https://qainsights.com/20-month-the-price-ceiling-every-ai-company-copied/ Comments URL: https://news.ycombinator.com/item?id=48962535 Points: 1 # Comments: 0
1.0.0
Versioning Note Since this should be the last release of the classic MCP Inspector, it is finally being promoted to 1.0.0, primarily so that it can be referred to as the V1 implementation. It fixes outstanding securit…
I asked a psychiatrist I know about the effects of AI
Article URL: https://teadrivendevelopment.substack.com/p/i-asked-a-psychiatrist-i-know-about Comments URL: https://news.ycombinator.com/item?id=48962463 Points: 1 # Comments: 0
Show HN: Graph Context Engine for Reliable AI
Been working on this for the last 2 years. There is a video on the site explaining how it works. There is also a production app powered by this engine. Give it a try, I'm happy to answer any questions. Comments URL: h…
Ada: An AI business intelligence software from CSV and Excel(yes LLMs but more)
Article URL: https://github.com/saineshnakra/automated-data-analyst Comments URL: https://news.ycombinator.com/item?id=48962405 Points: 1 # Comments: 0
Neil DeGrasse Tyson and Jaron Lanier on the AI Illusion
Article URL: https://www.youtube.com/watch?v=a_ZKYH8v_do Comments URL: https://news.ycombinator.com/item?id=48962160 Points: 3 # Comments: 0
One probable outcome of an open-weight-model-dominant world is full AI communism
Article URL: https://xcancel.com/deanwball/status/2078133895766114412?s=46 Comments URL: https://news.ycombinator.com/item?id=48961988 Points: 3 # Comments: 2
Databricks hits $188B valuation, extending its run as AI’s favorite second act
Article URL: https://techcrunch.com/2026/07/17/databricks-hits-188b-valuation-extending-its-run-as-ais-favorite-second-act/ Comments URL: https://news.ycombinator.com/item?id=48961957 Points: 3 # Comments: 0
FunASR llama.cpp runtime v0.1.7
Prebuilt self-contained binaries for the FunASR llama.cpp / GGUF runtime: SenseVoice, Paraformer and Fun-ASR-Nano with built-in FSMN-VAD. Download the default quantized model with bash download-funasr-model.sh (the he…
IBM CEO Arvind Krishna Has Nowhere to Hide from AI
Article URL: https://www.wsj.com/tech/ibm-ceo-arvind-krishna-has-nowhere-to-hide-from-ai-c9ff290f Comments URL: https://news.ycombinator.com/item?id=48961633 Points: 2 # Comments: 0
Jack Conte: Why I'm (sort of) not worried about AI [video]
Article URL: https://www.youtube.com/watch?v=17_HcR95YBc Comments URL: https://news.ycombinator.com/item?id=48961467 Points: 1 # Comments: 0
Show HN: Medows – an AI clinical workspace for doctors on ward rounds
Article URL: https://www.medows.ai/demo Comments URL: https://news.ycombinator.com/item?id=48961435 Points: 1 # Comments: 2
Spotify Deleted 75M AI-Generated Tracks – and It's Not Done Yet
Article URL: https://www.gadgetreview.com/spotify-deleted-75-million-ai-generated-tracks-and-its-not-done-yet Comments URL: https://news.ycombinator.com/item?id=48961347 Points: 1 # Comments: 0
StartupForge AI – Generate a startup blueprint from an idea
Article URL: https://b07ada739af0b4a7c99daf84e96b066a.ctonew.app/ Comments URL: https://news.ycombinator.com/item?id=48961254 Points: 1 # Comments: 0
Free AI Harness Profiler – What tf did Fable do with your tokens
Article URL: http://profiler.getrekon.com/ Comments URL: https://news.ycombinator.com/item?id=48961169 Points: 1 # Comments: 0
v3.0.0
What's new in 3.0.0 (2026-07-18) These are the changes in inference v3.0.0. New features FEAT: Switch default frontend to Next.js app by @OliverBryant in #5111 feat: (UI) API Key, User Management, refreshToken, change…
AI Karma Tracker
Article URL: https://ai-karma-tracker.github.io/ Comments URL: https://news.ycombinator.com/item?id=48961030 Points: 4 # Comments: 0
Kimi: Threat or menace?
Chinese company Moonshot AI released a new version of its Kimi model this week, prompting concern about "full AI communism."
White House cybersecurity clearinghouse to patch software flaws by AI
Article URL: https://www.politico.com/news/2026/07/14/white-house-launches-gold-eagle-cybersecurity-clearinghouse-to-patch-software-flaws-discovered-by-ai-00998011 Comments URL: https://news.ycombinator.com/item?id=48…
Show HN: A pre-launch scanner for AI-built apps (it scores itself 60/100)
Article URL: https://shippingszn.com/ Comments URL: https://news.ycombinator.com/item?id=48960665 Points: 2 # Comments: 0
OpenAI Strategic Lead Defines Open-Source AI as Dystopian Hellscape
Article URL: https://twitter.com/deanwball/status/2078133895766114412 Comments URL: https://news.ycombinator.com/item?id=48960571 Points: 2 # Comments: 0
Overtraining as the path to human-like AI
Article URL: https://www.seangoedecke.com/overtraining-as-the-path-to-human-like-ai/ Comments URL: https://news.ycombinator.com/item?id=48960118 Points: 13 # Comments: 3
SQLite Query Explainer
Tool: SQLite Query Explainer Julia Evan's, in Learning a few things about running SQLite : Maybe one day I’ll learn to read a query plan. Big same.... which inspired me to have Fable build this interactive explain too…
火速上架!空降即登顶Arena,Kimi K3当日登陆「模型广场」
点击查看原文>
v8.4.101 - Add Hailo pose, OBB, and classification export and inference for YOLOv8 and YOLO11 (#25276)
🌟 Summary 🚀 Ultralytics v8.4.101 expands Hailo deployment support across nearly the full YOLO task range and strengthens NDJSON dataset reliability, Platform validation insights, and documentation. 📊 Key Changes 🧠…
viable/strict/1784403717: [Dynamo] Fix self-referential bw_compiler on backend re-entry (#189325)
Why When one aot_autograd backend compiles a second, distinct graph, AotAutograd.__call__ re-enters with bw_compiler already wrapped. The _is_wrapped_bw_compiler branch then ran bw_compiler.compiler_fn = bw_compiler ,…
ciflow/trunk/21013: Fuse consecutive clamps into a single clamp (#21013)
Summary: ConvertToClampPass normalizes hardtanh, relu and relu6 into aten.clamp. Chains like HardTanh -> ReLU therefore become two adjacent clamps, the second of which is redundant. Add FuseConsecutiveClampsPass, whic…
trunk/391057a1ec7d1fefe518db8a729622706df04d6d: Skip test_workspace_runtime_integration (failing on main) (#190101)
Disables TestNVUniversalGemmEpilogueFusion.test_workspace_runtime_integration, failing in CI and tracked/disabled via #188674 . Test Plan: test-only; the test is skipped. Authored with Claude. Pull Request resolved: #…
viable/strict/1784399734: Skip test_workspace_runtime_integration (failing on main) (#190101)
Disables TestNVUniversalGemmEpilogueFusion.test_workspace_runtime_integration, failing in CI and tracked/disabled via #188674 . Test Plan: test-only; the test is skipped. Authored with Claude. Pull Request resolved: #…
trunk/90847591e411bd639b963b2d6545a1f3c2faa76a: [Docs] Clarify norm docs for complex inputs (#190381)
fixes: #136452 docstrings for linalg.norm and linalg.matrix_norm contain a confusing contradiction between the dtype kwarg and the return for complex dtypes. This is clarified by simply stating that the input is cast.…
viable/strict/1784397970: [Docs] Clarify norm docs for complex inputs (#190381)
fixes: #136452 docstrings for linalg.norm and linalg.matrix_norm contain a confusing contradiction between the dtype kwarg and the return for complex dtypes. This is clarified by simply stating that the input is cast.…
0.144.6
Bug Fixes Refreshed bundled instructions for GPT-5.6 Sol, Terra, and Luna, and corrected their context windows to 272,000 tokens. ( #33972 , #34009 ) Changelog Full Changelog: rust-v0.144.5...rust-v0.144.6 #33972 Back…
b10068
model: rotate injected K/V cache for DFlash ( #25823 ) dflash: rotate injected K/V cache when using K/V quantization Update src/models/dflash.cpp Co-authored-by: Georgi Gerganov ggerganov@gmail.com clearer format remo…
v1.2.3
This release introduces elliptic friction cone with high-impedance option to accurately model static function. In addition, the convergence of the constraint solver under fp32 accuracy has been improved. New Features…
v3.44.0
3.44.0 (2026-07-17) Full Changelog: v3.43.0...v3.44.0 Features api: /organization/projects/{project_id}/service_accounts/{service_account_id}/api_keys" endpoint ( 28d2a6a ) api: manual updates ( a531d4a ) api: manual…
trunk/63d4ffc9b63c5f97af9c01d8bf24014de517561c
Remove unused USE_OPENCL, USE_OBSERVERS and USE_SNPE build options (#...
viable/strict/1784396290
Remove unused USE_OPENCL, USE_OBSERVERS and USE_SNPE build options (#...
b10067
llama-quant : exclude i32 ffn_gate_tid2eid routing table from quantization ( #25787 ) DeepSeek-V4's ffn_gate_tid2eid tensor is an i32 token-id -> expert-id index table, not weights. It was never added to the name-base…
中国 AI 产业核心要素出海发展白皮书
点击查看原文>
viable/strict/1784394527: Fix typos in comments and docstrings across torch (#190249)
Correct spelling and grammar mistakes in code comments and docstrings throughout the torch package. These are documentation-only changes with no functional impact, covering typos such as "TensorQeue" to "TensorQueue",…
China's new World Artificial Intelligence Cooperation Organization is President Xi's clearest play yet for a parallel AI order
At the World AI Conference in Shanghai, Xi Jinping announced 5,000 AI training slots for Global South countries and the launch of the "World Artificial Intelligence Cooperation Organization." Cooperation centers with…
Open-weight models now match frontier cyber performance from just four months ago at a fraction of the cost
The British AI Security Institute warns that open-weight models like GLM-5.2 and DeepSeek V4-Pro now trail closed frontier models in cyber capabilities by four to seven months. At the start of 2025, the gap was still…
为什么 AI Agent 拿到数据却不会推理?可观测对象图语义层的设计与开源实践|AICon深圳
点击查看原文>
AI 智能体算力消耗过快,传统账单风控跟不上速度
点击查看原文>
豆包视频通话背后,火山引擎重构 Agent 时代多模态传输底座
点击查看原文>
不止于 4K,火山引擎画质增强让视频从清晰走向细腻
点击查看原文>
The Pentagon's new AI playbook treats slow adoption as a bigger risk than imperfect alignment
The US Department of the Navy has signed a strategy to "weaponize" data and AI and build an "AI-first" fleet. Large language models would run directly on warships, and an AI war council would prioritize mission scenar…