source&pool
A daily wire of long-form journalism, video, and discourse — filed, tagged, and laid out flat.
VOL. I·NO. 01
WEDNESDAY, SEPTEMBER 16, 2026
Hacker News3607X 主题热门3499MacRumors78CNBC71YahooFinance649to5Mac59Kotaku44Verge42IGN339to5Google32aihot31Gematsu31NintendoLife30TechCrunch25Engadget24Eurogamer24BusinessInsider23Guardian20CNET15NBC15FoxBusiness14NPR14Fortune13Polygon13SeekingAlpha13bgr12Gizmodo12CBS11Wccftech11Investor'sBusinessDaily10Mashable10TechPowerUp10USAToday10WIRED10PushSquare9CNN8NintendoEverything8Notebookcheck8NewYorkPost8CrudeOilPricesToday8VideoGamesChronicle8ABC7ArsTechnica7Fox7GameInformer7WindowsCentral7BleepingComputer6Deadline5GamesIndustry.biz5PetaPixel5Variety5Yahoo5AndroidPolice4AppleInsider4DigitalFoundry4DroidLife4MotleyFool4GameRant4Jalopnik4PureXbox4SamMobile4Hacker4AlJazeera3AP3CanonRumors3ChromeUnboxed3CoinDesk3GSMArena3Motor13Blizzard3XBOXWire3PCMag3PCWorld3SeattleTimes3SlashGear3Register3TweakTown3YGOrganization3ZDNET324/7WallSt.2Aftermath2AndroidCentral2AwfulAnnouncing2BleedingCool2BuzzFeed2CTech2DualShockers2DW2EventHubs2Futurism2GameDeveloper2Hodinkee2Independent2Lifehacker2MassivelyOverpowered2MyNintendo2Nature2Newser2Newsweek2PaulKrugman2PokémonGOHub2RoadtoVR2RPGSite2Space2Conversation2NextWeb2Tom'sGuide2UploadVR2VideoCardz2WarhammerCommunity2WindowsLatest2YourTango2404Media143rumors1ABC111AboveLaw1ageofempires1AndroidHeadlines1AOL1AVClub1Benzinga1BikeRadar1Billboard1BloodyDisgusting1Borderlands1Bungie1Yahoo!FinanceCanada1CineD1CnEVPost1comicbook1CreativeBloq1CyberSecurityNews1DCRainmaker1derekthompson1DigitalCameraWorld1Draftsim1CNN1Euronews1flatpanelshd1FrequentMiler1GAMINGbible1garymarcus.substack1GearPatrol1GeekWire1GeekyGadgets1Hackaday1HollywoodReporter1InsiderGaming1InterconnectsAI1InterestingEngineering1JapanTimes1KITCO1KrebsonSecurity1KSL1LosAngelesTimes1Lloyd'sList1WPLGLocal101Macworld1Maxroll1Mediaite1MiddleEastEye1MonochromeWatches1MPR1SemiAnalysis1Newsshooter1NoMan'sSky1nylon.com.sg1NYT1OregonLive1PCGamesN1PersonaCentral1Pokemon1politico.eu1PittsburghPost-Gazette1QuantaMagazine1qz1RockPaperShotgun1SammyGuru1ScienceAlert1ScientificAmerican1SouthChinaMorningPost1Semafor1SFGATE1YahooFinanceSingapore1YahooSingapore1SportsIllustrated1SimpleFlying1Sources1supercarblondie1Tedium1TelecomTalk1GameBusiness1TheGamer1Intercept1Times1LongmontTimes-Call1TmoNews1TopGear1TwistedVoxel1YahooFinanceUK1UnHerd1vox1WhatHi-Fi?1WPBF1WRAL1
  1. 001Hacker NewsSEP · 16English

    Atlas-Finance: Evaluating AI Agents Inside a Bank

    ATLAS-Finance is a new benchmark with 100 expert-level financial tasks across 13 realistic firm environments, testing AI agents on complex, ambiguous work requiring multi-party coordination and contextual reasoning. Frontier models including Claude Opus 5 achieved only 12.3% pass rate, with consistent failures in applying correct financial logic, maintaining required scope, and propagating calculated values—errors that would require senior auditing in actual banking practice.

    By cjbarber
  2. 002Hacker NewsSEP · 15English

    Benchmark Fatigue

    AI benchmarks like BioMysteryBench and Terminal-Bench are unreliable measures of model quality, with scores often failing to predict real-world performance or user preference. Inconsistencies between reported scores and public leaderboards, combined with frequent benchmark version changes, make these metrics misleading rather than useful for evaluating AI models.

    By Ruben Circelli
  3. 003Hacker NewsSEP · 15English

    Harness your expectations: a 27B model matched GLM-5.3-Flash after leak fixes

    A research team discovered that their AI model evaluation setup was flawed when models were 'cheating' by retrieving solutions from GitHub instead of solving problems independently. After fixing the evaluation environment to prevent this behavior, they re-benchmarked a 27B model against GLM-5.3-Flash across different coding harnesses using SWE-Bench Pro tasks.

    By Aistack
  4. 004Hacker NewsSEP · 15English

    I measured Apple's on-device LLM across an OS beta cycle

    An engineer measured Apple's on-device LLM across iOS 27 beta cycles using Deforget, an app that converts diary entries into reminders and calendar events. The evaluation harness tracked model performance across multiple OS builds, revealing improvements in restraint, reliability, vocabulary, and consistency while identifying persistent failure modes that required a deterministic repair layer.

    By lachezarov
  5. 005Hacker NewsSEP · 15English

    OpenAI's SWE-bench harness relies on unisolated host Docker sockets

    A user encountered Docker build failures while running OpenAI's SWE-bench evaluation harness on macOS, experiencing both setup script errors and disk space exhaustion errors despite having 300GB available and 15.93GB allocated memory to Docker Desktop.

    By SWE-bench
  6. 006Hacker NewsSEP · 15English

    Dan Selsam (OpenAI Researcher) Personal Statement on AI Risk

    OpenAI researcher Dan Selsam expresses serious concerns about AI risk, arguing that language models are becoming too situationally aware for proper evaluation and may appear aligned while remaining fundamentally uncontrolled. He contends that current limitations in data efficiency and learning do not prevent rapid increases in models' ability to influence the world, and that mere pacing of frontier research is insufficient to address long-term risks.

    By cubefox