跳到主要內容
AI News HubLIVE

AI 新聞即時情報

即時監測

今天 AI 世界最重要的變化

來源目錄收錄 105 個來源 · 本頁最新發布時間 2026-08-28 10:38 UTC+8。

部分報導的翻譯與分析尚未完成,已標示來源內容。精選優先顯示處理完成的報導。

即時監測

即時更新

即時追蹤可信來源,保留出處、權限和站內閱讀模式,把噪音壓成可讀情報。

即時更新

重設

2026-08-28

待翻譯:Show HN: ChessRabbit – The AI Chess Analysis Platform

AI 服務暫時不可用,以下為來源摘要,待恢復後補全翻譯:Hi HN! I wanted to share with you a chess analysis tool that I was building for the past week. First of all I want to explain what is the problem that I'm trying to solve: Chess Engines such as stockfish are superior fo…

Hacker News AIAgent / 創業融資來源內容 · 翻譯待補全待翻譯:Show HN: ChessRabbit – The AI Chess Analysis Platform
待翻譯:Shai-Hulud was the best thing to happen to supply chain security

AI 服務暫時不可用,以下為來源摘要,待恢復後補全翻譯:Shai-Hulud was the best thing to happen to supply chain security We might Shai-Hulud to thank for convincing the community to use Trusted publishing Charlie Eriksen Published on: Aug 24, 2026 Last updated on: Aug 26, 20…

Hacker News AIAgent / 創業融資來源內容 · 翻譯待補全待翻譯:Shai-Hulud was the best thing to happen to supply chain security
待翻譯:Ask Me Twice – A longitudinal archive of AI chatbot responses

AI 服務暫時不可用,以下為來源摘要,待恢復後補全翻譯:Ask Me Twice Ask Me Twice A longitudinal archive of AI chatbot responses: a curated, evolving set of questions is put to a wide range of AI models every day, and the responses are recorded so you can compare how models…

Hacker News AI模型 / Agent來源內容 · 翻譯待補全待翻譯:Ask Me Twice – A longitudinal archive of AI chatbot responses
待翻譯:Alphabet stock sheds $700B as AI bills climb

AI 服務暫時不可用,以下為來源摘要,待恢復後補全翻譯:Alphabet’s stock is down more than 15% from its May peak, wiping out roughly $700 billion in value, as concerns rise over its mounting AI bills and failure to produce a decisive technological breakthrough. After leading…

Hacker News AI工具來源內容 · 翻譯待補全待翻譯:Alphabet stock sheds $700B as AI bills climb
待翻譯:Anthropic pushes into physical world with standard to help agents run machines

AI 服務暫時不可用,以下為來源摘要,待恢復後補全翻譯:Anthropic pushes into physical world with new standard to help AI agents operate machines Skip Navigation Anthropic announced the Model Hardware Standard, or MHS, a new interface that will make it simpler for AI agents…

Hacker News AIAgent來源內容 · 翻譯待補全待翻譯:Anthropic pushes into physical world with standard to help agents run machines
待翻譯:Black Box: episode 5 – The white mask – podcast

AI 服務暫時不可用,以下為來源摘要,待恢復後補全翻譯:Revisited: Guardian journalist Michael Safi looks into the world of artificial intelligence, exploring the dangers and promises it holds for society Today in Focus is on a summer break and will be back with new episodes from 1 September. In the meantime, we are bringing you season one of Black Box, before the launch of season two in early September. This episode was first broadcast on 18 March 2024. In January 2020, Robert Williams was arrested by Detroit police for a crime he had not committed. The officers were acting on a tipoff, but not from a witness or informant. In fact, not from a person at all. Continue reading...

The Guardian AI工具來源內容 · 翻譯待補全待翻譯:Black Box: episode 5 – The white mask – podcast
待翻譯:Supporting Thailand’s next generation of AI startups

AI 服務暫時不可用,以下為來源摘要,待恢復後補全翻譯:OpenAI and Thailand’s MHESI launch an eight-week accelerator helping 10 health, wellness, and education startups turn AI prototypes into trusted products.

OpenAI News創業融資來源內容 · 翻譯待補全待翻譯:Supporting Thailand’s next generation of AI startups
待翻譯:AI-Powered Tilichu VST/AU Synth Released by Butterlamp Audio

AI 服務暫時不可用,以下為來源摘要,待恢復後補全翻譯:Butterlamp Audio has introduced a new wavetable synthesizer built for classic, knob-by-knob sound design with depth on every layer. Tilichu features two morphing wavetable oscillators, granular and spectral sample engin…

Hacker News AI工具來源內容 · 翻譯待補全待翻譯:AI-Powered Tilichu VST/AU Synth Released by Butterlamp Audio
待翻譯:Show HN: A public feed of website changes

AI 服務暫時不可用,以下為來源摘要,待恢復後補全翻譯:Monity.ai — AI Website Change Monitoring & Intelligence AI-powered website changes monitoring & alerts Loading... Preparing your AI workspace Connecting monitors, alerts, and agents

Hacker News AIAgent來源內容 · 翻譯待補全待翻譯:Show HN: A public feed of website changes
待翻譯:Fast, fault-tolerant PyTorch training on AI Runtime

AI 服務暫時不可用,以下為來源摘要,待恢復後補全翻譯:At scale, your training efficiency is determined by a single metric: "goodput", the...

Databricks Blog研究來源內容 · 翻譯待補全待翻譯:Fast, fault-tolerant PyTorch training on AI Runtime
待翻譯:Awareness Local: local-first memory for AI coding agents (96% R5 on LongMemEval)

AI 服務暫時不可用,以下為來源摘要,待恢復後補全翻譯:Notifications You must be signed in to change notification settings Fork 1 Star 8 BranchesTags Open more actions menu Latest commit History 79 Commits 79 Commits Folders and files NameName Last commit message Last commi…

Hacker News AIAgent / 研究來源內容 · 翻譯待補全待翻譯:Awareness Local: local-first memory for AI coding agents (96% R5 on LongMemEval)
待翻譯:Yet Another AI Builder But

AI 服務暫時不可用,以下為來源摘要,待恢復後補全翻譯:wanno: AI web & app builder Describe your idea. It's live. Only for iPhone Free · In‑App Purchases · Designed for iPhone. Not verified for macOS. Age Rating 4+ Years Category Productivity Developer FEDERICO GUILLERMO CA…

Hacker News AI工具來源內容 · 翻譯待補全待翻譯:Yet Another AI Builder But
待翻譯:Anthropic previews MHS standard for AI agents that operate machines

AI 服務暫時不可用,以下為來源摘要,待恢復後補全翻譯:Anthropic PBC today previewed a standard that makes it easier for artificial intelligence agents to control machines such as microscopes. The Model Hardware Standard, or MHS, is the fruit of a collaboration between the Claude developer and medical research institute HHMI. Anthropic has so far only made the technology accessible to a limited number of […] The post Anthropic previews MHS standard for AI agents that operate machines appeared first on SiliconANGLE.

SiliconANGLE AIAgent / 研究來源內容 · 翻譯待補全待翻譯:Anthropic previews MHS standard for AI agents that operate machines
待翻譯:Terminal-Bench-Science: Evaluating AI agents on scientific research workflows

AI 服務暫時不可用,以下為來源摘要,待恢復後補全翻譯:Terminal-Bench-Science 0.1 Terminal-Bench-Science evaluates AI agents on workflows from researchers' own work. Scientists, not model developers or data vendors, set the bar for scientific capability in AI. Terminal-Benc…

Hacker News AIAgent / 研究來源內容 · 翻譯待補全待翻譯:Terminal-Bench-Science: Evaluating AI agents on scientific research workflows
待翻譯:Putting Task Expertise into RL Achieves Performance on Text-to-SQL

AI 服務暫時不可用,以下為來源摘要,待恢復後補全翻譯:Many industries rely on relational databases that are queried with SQL. Most SQL is machine-written, but humans alone likely write billions of custom SQL queries each monthBased on our internal estimates and publicly av…

Hacker News AIAgent / 研究來源內容 · 翻譯待補全待翻譯:Putting Task Expertise into RL Achieves Performance on Text-to-SQL
待翻譯:GLM-5.3 vs. GLM-5.3 Flash on DeepSWE: Cost, Coding, and Routing

AI 服務暫時不可用,以下為來源摘要,待恢復後補全翻譯:We ran 900 DeepSWE rollouts on GLM-5.3 and GLM-5.3 Flash. Flash gives up 5.6 points of pass@1 at 17x lower cost, and only 2.6 points at pass@4.

Together AI Blog模型 / 研究來源內容 · 翻譯待補全待翻譯:GLM-5.3 vs. GLM-5.3 Flash on DeepSWE: Cost, Coding, and Routing
待翻譯:Agent Seer: Synthesizing Scenarios from Specification Understanding

AI 服務暫時不可用,以下為來源摘要,待恢復後補全翻譯:Evaluating AI agents that use external tools requires realistic test scenarios that capture how practitioners compose tools and iterate across conversation turns. Constructing such scenarios by hand demands deep domain expertise, does not scale across tool ecosystems, and produces static benchmarks that cannot track evolving APIs. We observe that tool specifications—function names, natural-language descriptions, and typed parameter schemas—already encode sufficient semantic information to synthesize realistic evaluation scenarios without manual curation or live tool execution. Agent Seer…

Apple Machine Learning ResearchAgent / 研究來源內容 · 翻譯待補全待翻譯:Agent Seer: Synthesizing Scenarios from Specification Understanding
待翻譯:LLMs Are Not (Consistently) Bayesian: Quantifying Internal (In)consistencies of LLMs’ Probabilistic Beliefs

AI 服務暫時不可用,以下為來源摘要,待恢復後補全翻譯:Modern AI systems are being deployed in complex domains such as medicine, science, and law, where there is often not a single correct answer given the observed evidence. Such systems must be able to represent and update uncertain beliefs about the world as new evidence arrives to make rational decisions. We introduce the novel technique of studying LLMs as information processing rules and utilize the information processing gap—the deviation from Bayes updates—to study the internal (in)consistencies of how LLMs update their probabilistic beliefs from evidence. Our extensive experiments evaluate…

Apple Machine Learning Research模型 / 研究來源內容 · 翻譯待補全待翻譯:LLMs Are Not (Consistently) Bayesian: Quantifying Internal (In)consistencies of LLMs’ Probabilistic Beliefs
待翻譯:XPress: Parallel Refinement for Diffusion Drafters in Speculative Decoding

AI 服務暫時不可用,以下為來源摘要,待恢復後補全翻譯:XPress: Parallel Refinement for Diffusion Drafters in Speculative Decoding August 2, 2026 12 minutes Case study. A real GSM8K prompt decoded three ways under the same timing setup: autoregressive, the dFlash drafter alo…

Hacker News AI模型 / 研究來源內容 · 翻譯待補全待翻譯:XPress: Parallel Refinement for Diffusion Drafters in Speculative Decoding
待翻譯:AI Finds Critical Flaw in Bitcoin Lightning, Devs Issue Emergency Warning

AI 服務暫時不可用,以下為來源摘要,待恢復後補全翻譯:In brief Core Lightning confirmed that several AI-generated security reports identified real flaws. The project told operators to verify and install its forthcoming update promptly. Operators who cannot upgrade should u…

Hacker News AI工具來源內容 · 翻譯待補全待翻譯:AI Finds Critical Flaw in Bitcoin Lightning, Devs Issue Emergency Warning
待翻譯:Unlimited AI calls, done sustainably

AI 服務暫時不可用,以下為來源摘要,待恢復後補全翻譯:Unlimited, sustainably. Switch once. Keep building. 0code is sustainably unlimited and will always be free. Why switch to 0code? One request, no SDK. Call GET /api/chat/ from any language, script, or prototype. During h…

Hacker News AI工具來源內容 · 翻譯待補全待翻譯:Unlimited AI calls, done sustainably
待翻譯:Three UK airports hit by cyber-attack with data of 8.7M customers accessed

AI 服務暫時不可用,以下為來源摘要,待恢復後補全翻譯:Manchester, London Stansted and East Midlands airports have been hit by a cyber-attack in which hackers accessed the data of about 8.7 million customers. The incident involved data related to “car park, lounge and fast-…

Hacker News AI工具來源內容 · 翻譯待補全待翻譯:Three UK airports hit by cyber-attack with data of 8.7M customers accessed
待翻譯:Build agentic creative workflows with Amazon Quick and fal

AI 服務暫時不可用,以下為來源摘要,待恢復後補全翻譯:Creative teams produce more assets than ever, but fragmented tools and manual context transfer slow production. This post shows how to build a reusable agent harness with Amazon Quick and fal, connected through the Model Context Protocol (MCP), using two hands-on workflows: an eight-panel storyboard and a music-video concept prototype.

AWS Machine Learning BlogAgent / 研究來源內容 · 翻譯待補全待翻譯:Build agentic creative workflows with Amazon Quick and fal
待翻譯:Breaking Claude Code Opus 5 Auto Mode

AI 服務暫時不可用,以下為來源摘要,待恢復後補全翻譯: Breaking Claude Code Opus 5 Auto Mode Anthropic are putting a great deal of faith in Claude Code's auto mode for protecting their coding agent users against prompt injection attacks. They recently made that the default and have made bold claims about its effectiveness. Johann Rehberger is one of the most credible prompt injection researchers active today. He found an attack against auto mode which he claims works 80% of the time, by tricking Claude Code into downloading and uncompressing a zip archive, then executing code that imports base64 without noticing that this will import and execute a local struct.py file extracted from the archive. In a few cases auto mode directly prevented the agent from preventing harmful code from continuing to execute! In a few…

Simon Willison's WeblogAgent / 模型 / 研究來源內容 · 翻譯待補全待翻譯:Breaking Claude Code Opus 5 Auto Mode
待翻譯:Show HN: Beating GPT5.5-xhigh for Coding agent security with SLMs and IRM

AI 服務暫時不可用,以下為來源摘要,待恢復後補全翻譯:Coding agents craft arbitrary code so securing them is more complicated than red-teaming. We post trained a cyber-security small llm, changed how it reasons and supplemented our controls using program analysis technique…

Hacker News AIAgent / 模型 / 研究來源內容 · 翻譯待補全待翻譯:Show HN: Beating GPT5.5-xhigh for Coding agent security with SLMs and IRM
待翻譯:ThunderPhone

AI 服務暫時不可用,以下為來源摘要,待恢復後補全翻譯:Discussion | Link

Product Hunt AIAgent來源內容 · 翻譯待補全待翻譯:ThunderPhone
待翻譯:Consumer-focused AI assistant startup Instinct reportedly raising $250M

AI 服務暫時不可用,以下為來源摘要,待恢復後補全翻譯:Instinct, the developer of an artificial intelligence assistant popular among Silicon Valley tech workers, is reportedly raising $250 million in funding. The company told the Wall Street Journal on Wednesday that the round is being co-led by Index Ventures and Benchmark. It’s set to value Instinct at $2.5 billion. The startup previously raised $100 million […] The post Consumer-focused AI assistant startup Instinct reportedly raising $250M appeared first on SiliconANGLE.

SiliconANGLE AI研究 / Agent來源內容 · 翻譯待補全待翻譯:Consumer-focused AI assistant startup Instinct reportedly raising $250M
待翻譯:Show HN: Make apps in seconds inside of sandbox and share them with a link

AI 服務暫時不可用,以下為來源摘要,待恢復後補全翻譯:deenesjakoruzh This server lets agents manage persistent, forkable cloud development environments over MCP. Check compute credits – see the account's remaining compute-credit balance. Create environments – spin up a per…

Hacker News AIAgent來源內容 · 翻譯待補全待翻譯:Show HN: Make apps in seconds inside of sandbox and share them with a link
待翻譯:My $120 GE smart scale's body fat is just 1.5 × BMI − 17.5

AI 服務暫時不可用,以下為來源摘要,待恢復後補全翻譯:GE scales are a scam and math theater I exported 3 months of readings from my GE CS10H smart scale. The body fat percentage comes from a formula: fat % = 0.434 × weight − 17.5. Every one of 37 readings fits within 0.1 p…

Hacker News AI研究來源內容 · 翻譯待補全待翻譯:My $120 GE smart scale's body fat is just 1.5 × BMI − 17.5