[{"data":1,"prerenderedAt":419},["ShallowReactive",2],{"project-45265831":3},{"id":4,"name":5,"line":6,"line_source":7,"title":8,"url":9,"domain":10,"maker":11,"posted_at":12,"year":13,"points":14,"comments":15,"rank_in_year":16,"shown_in_year":17,"top":18,"tier":19,"category":20,"subcategory":21,"format":22,"keywords":23,"reception":28,"sentiment":29,"stance_mix":30,"scored_comments":35,"aspects":36,"strip":47,"post_text":48,"what":49,"said":71,"maker_in_thread":205,"shape":216,"grid":219,"attempts":382,"same_week":383,"maker_shows":256,"maker_hidden":411,"reader_version":412,"card":413},"45265831","AI Code Detector","detect AI-generated code with 95% accuracy","title","Show HN: AI Code Detector – detect AI-generated code with 95% accuracy","https:\u002F\u002Fcode-detector.ai\u002F","code-detector.ai","henryl",1758046736,2025,72,65,989,3837,26,"notable","dev_tools","ai code detection","web_app",[24,25,26,27],"ai code detector","detect ai generated code","code provenance","engineering analytics","mixed",-0.281,{"positive":31,"neutral":32,"mixed":33,"critical":34},5,19,3,14,41,{"performance":37,"privacy":40,"idea":41,"docs":43,"execution":44,"design":45,"pricing":46},{"positive":38,"critical":39},0,12,{"positive":38,"critical":38},{"positive":33,"critical":42},1,{"positive":42,"critical":38},{"positive":42,"critical":42},{"positive":38,"critical":38},{"positive":38,"critical":38},[42,42,33,42,33,38,38,31,31,42,33,42,38,38,42,38,42,33,33,33,42,38,33,42,33,33,42,33,31,38,42,38,42,42,42,33,33,33,31,33],"Hey HN,\n\nI’m Henry, cofounder and CTO at Span (https:\u002F\u002Fspan.app\u002F). Today we’re launching AI Code Detector, an AI code detection tool you can try in your browser.\n\nThe explosion of AI generated code has created some weird problems for engineering orgs. Tools like Cursor and Copilot are used by virtually every org on the planet – but each codegen tool has its own idiosyncratic way of reporting usage. Some don’t report usage at all.\n\nOur view is that token spend will start competing with payroll spend as AI becomes more deeply ingrained in how we build software, so understanding how to drive prof",{"problem":50,"audience":56,"pricing":56,"motivation":57,"time_to_build":56,"stack":61,"openness":68,"stage":69,"team_size":70},{"value":51,"source":52,"evidence":53},"Engineering orgs lack visibility into how much of their shipped code is AI-generated.","maker",[54],{"id":55},"post",null,{"value":58,"source":52,"evidence":59},"Customers ask for visibility into AI-generated code; token spend is becoming a major cost to manage.",[60],{"id":55},[62],{"value":63,"source":52,"evidence":64},"span-detect-1 model, unified across languages",[65],{"id":66,"by":11,"text":67},"45266646","It's probably impossible to detect ALL languages without training for them specifically, but there's good generalization happening. Our model is a unified model rather than a…","unknown","launched","company",{"praise":72,"objections":92,"requests":142,"questions":174,"alternatives":204},[73],{"value":74,"source":75,"evidence":76},"Impressive results and useful for engineering leaders.","commenter",[77,81,85,89],{"id":78,"by":79,"text":80},"45266515","mechen","As a leader this is actually really neat - going to give it a spin",{"id":82,"by":83,"text":84},"45266914","ht-syseng","Just tried. Actually quite impressed with how well it works. I avoid using AI to write code, I'm a little worried that the existence of detection tools like this will lead people…",{"id":86,"by":87,"text":88},"45268858","khanna_ayush","My engineers didn’t know how much they used AI for vibe coding until I used Span. Can confirm we were all left with jaws on the floor. Now re-thinking my hiring plan for the next…",{"id":90,"by":79,"text":91},"45266560","Just tried it out and it works :mind-blown:",[93,104,119,134],{"value":94,"source":75,"evidence":95},"95% accuracy yields too many false positives for consequential use.",[96,100],{"id":97,"by":98,"text":99},"45269441","kittikitti","A 95% accuracy is very low for this type of thing. People use this to enact administrative consequences. People's lives are ruined and 5% is too high of a false positive rate.…",{"id":101,"by":102,"text":103},"45265960","bigyabai","I can detect AI-generated code with 100% accuracy, provided you give me an unlimited budget for false positives. It's a bit of a useless metric.",{"value":105,"source":75,"evidence":106},"Easy to evade by editing AI code to look human-written.",[107,111,115],{"id":108,"by":109,"text":110},"45266824","mendeza","I feel like code fed into this detector can be manipulated to increase false positives. The model probably learns patterns that are common in generated text (clean comments, AI…",{"id":112,"by":113,"text":114},"45266175","jensneuse","Could I use this to iterate over my AI generated code until it's not detectable anymore? So essentially the moment you publish this tool it stops working?",{"id":116,"by":117,"text":118},"45268483","jakderrida","What if I just modify the code to misspell things that no AI would misspell?",{"value":120,"source":75,"evidence":121},"Gave wrong results on human and AI test code.",[122,126,130],{"id":123,"by":124,"text":125},"45272831","yamalight","It told me my ~10 year old js project was 50% AI generated. Yeah, this is more or less the same as \"AI text detector\" stuff that won't work reliably (but people who don't…",{"id":127,"by":128,"text":129},"45268272","simanyay","Very cool! I wonder if it performs differently on actual “production” code versus random tests? I opened ChatGPT, typed a random non-sensical prompt, copy-pasted the response[1]…",{"id":131,"by":132,"text":133},"45269087","thirdacc","And false negatives. I just pasted 100% AI generated code and it told me it's only 40% AI written.",{"value":135,"source":52,"evidence":136},"Only supports TypeScript and Python; other languages missing.",[137,138],{"id":55},{"id":139,"by":140,"text":141},"45266457","mannicken","Only Python, TypeScript and JavaScript? Well there go my vibe-coded elisp scripts. I guess it's impossible (or really hard) to train a language-agnostic classifier. Reference,…",[143,157,167],{"value":144,"source":75,"evidence":145},"Support for Java, Ruby, C# and C.",[146,150,154],{"id":147,"by":148,"text":149},"45266526","Alifatisk","Very cool piece of tech, I would suggest putting C on the priority list and then Java. Mainly because Unis and Colleges use one of them or both, so that would be a good use case",{"id":151,"by":152,"text":153},"45266564","johnsillings","Totally – we have support for Java, C#, and Ruby in the works. Edit: since you mentioned universities, are you thinking about AI detection for student work, e.g. like a…",{"id":155,"by":148,"text":156},"45267169","Glad to hear Ruby is being in the list as well! When it comes to the unis, I was thinking of both AI detection for student work. I mean like plagiarism checkers are common…",{"value":158,"source":75,"evidence":159},"CLI or CI\u002FCD integration.",[160,164],{"id":161,"by":162,"text":163},"45266406","icemanx","Would be amazing to have a CLI tool that detects AI generated code (even add it as part of CI\u002FCD pipelines). I'm tired of all the AI trash PRs",{"id":165,"by":152,"text":166},"45267162","It is possible to use this via the command line today. I'll ask Henry to have a look & comment here (or grab a demo and leave a note, we'll give you some more details).",{"value":168,"source":75,"evidence":169},"Option to skip comments when analysing code.",[170],{"id":171,"by":172,"text":173},"45267484","jftuga","I wonder if you could add a toggle to only examine only source and skip comments.",[175,186,200],{"value":176,"source":75,"evidence":177},"What is the pricing and is it B2B or B2C?",[178,182],{"id":179,"by":180,"text":181},"45268046","jimmydin7","what will the pricing be? i guess this is just a super early demo, I want to hear your pricing plan. Also, is this B2B or B2C?",{"id":183,"by":184,"text":185},"45266781","jjmarr","You're saying \"Understand and report on impact by AI coding tool\". How can you drill down into per-coding assistant usage? Also, what's the pricing?",{"value":187,"source":75,"evidence":188},"What are precision and recall, and what is the false positive rate?",[189,193,197],{"id":190,"by":191,"text":192},"45266092","samfriedman","Accuracy is a useless statistic: give us precision and recall.",{"id":194,"by":195,"text":196},"45266761","p0w3n3d","I wonder how many false positives it has",{"id":198,"by":11,"text":199},"45266171","Recall 91.5, F1 93.3",{"value":201,"source":75,"evidence":202},"Can it show usage per coding assistant?",[203],{"id":183,"by":184,"text":185},[],{"comments":206,"answered":207,"key":208},7,4,[209,210,213],{"id":198,"text":199},{"id":211,"text":212},"45266038","I'd argue that knowing AI generated code shipped into production is the first step to understanding the impact of AI coding assistants on velocity and quality. When paired with…",{"id":214,"text":215},"45266220","Ah - I misread: Recall 91.5, F1 93.3",{"commenters":35,"max_depth":207,"maker_replies":206,"first_comment_mins":217,"lifespan_hours":218},9,242,[220,222,224,227,230,234,238,240,244,248,251,253,255,258,260,261,263,266,270,273,275,279,282,283,285,288,289,292,295,296,298,300,302,304,307,309,310,312,313,315,319,321,324,326,330,332,334,336,338,340,342,345,348,350,351,354,356,359,361,365,367,369,373,376,378],[108,38,42,221,109,38],"project\u002Fcritical\u002Fperformance",[223,38,42,221,109,38],"45266884",[225,38,33,226,152,38],"45267144","maker\u002Fneutral\u002Fprivacy",[228,38,33,229,109,38],"45267344","project\u002Fneutral\u002Fidea",[231,38,42,232,233,38],"45267793","project\u002Fcritical\u002Fidea","runako",[235,38,38,236,237,38],"45267277","commenter\u002Fneutral\u002Fnone","nomel",[239,38,38,236,109,38],"45267373",[241,38,33,242,243,38],"45267036","project\u002Fneutral\u002Fperformance","bbsbb",[245,42,38,246,247,38],"45266841","other\u002Fneutral\u002Fnone","fancyfredbot",[249,42,38,246,250,38],"45266866","czbond",[252,42,38,246,247,38],"45267535",[254,42,38,246,152,38],"45266950",[161,256,31,257,162,38],2,"project\u002Fpositive\u002Fidea",[165,256,31,259,152,38],"maker\u002Fpositive\u002Fdocs",[190,33,42,221,191,38],[198,33,38,262,11,42],"maker\u002Fneutral\u002Fperformance",[264,33,33,262,265,38],"45266926","paradite",[267,33,33,268,269,38],"45266557","project\u002Fmixed\u002Fperformance","LPisGood",[271,33,42,221,272,38],"45266731","dymk",[274,33,38,236,269,38],"45266809",[276,33,38,277,278,38],"45269477","commenter\u002Fcritical\u002Fperformance","spott",[280,207,38,246,281,38],"45270518","ldl12345",[101,31,42,221,102,38],[211,31,38,284,11,42],"maker\u002Fpositive\u002Fidea",[286,31,38,236,287,38],"45266138","jfarina",[214,31,38,262,11,42],[139,290,42,291,140,38],6,"project\u002Fcritical\u002Fexecution",[293,290,33,294,152,38],"45266539","maker\u002Fneutral\u002Fexecution",[66,290,38,294,11,42],[297,206,33,262,172,38],"45266865",[299,206,33,262,152,38],"45267240",[171,206,33,301,172,38],"project\u002Fneutral\u002Fdesign",[123,303,42,221,124,38],8,[305,303,38,236,306,38],"45272905","TrueGeek",[308,303,38,236,124,38],"45273302",[147,217,33,229,148,38],[151,217,33,311,152,38],"maker\u002Fneutral\u002Fidea",[155,217,33,311,148,38],[127,314,42,221,128,38],10,[316,317,33,242,318,38],"45266158",11,"JohnFriel",[320,317,38,262,11,42],"45266216",[322,39,33,323,152,38],"45266177","maker\u002Fneutral\u002Fdocs",[97,325,42,221,98,38],13,[327,34,33,328,329,38],"45266956","project\u002Fmixed\u002Fidea","triwats",[331,34,33,262,152,38],"45267196",[86,333,31,257,87,38],15,[78,335,31,257,79,38],16,[337,335,38,236,152,38],"45266552",[112,339,42,221,113,38],17,[341,339,38,262,11,42],"45266192",[343,339,38,277,344,38],"45266606","polynomial",[346,339,38,246,347,38],"45266241","well_actulily",[194,349,42,221,195,38],18,[131,349,42,221,132,38],[352,32,33,229,353,38],"45270792","faangguyindia",[116,355,42,221,117,38],20,[179,357,33,358,180,38],21,"project\u002Fneutral\u002Fpricing",[82,360,33,328,83,38],22,[362,363,33,262,364,38],"45266164",23,"Ndotkess",[366,363,38,323,11,42],"45266181",[368,363,33,242,152,38],"45266190",[370,371,38,246,372,38],"45390703",24,"preyapatel",[90,374,31,375,79,38],25,"project\u002Fpositive\u002Fexecution",[183,18,33,377,184,38],"project\u002Fneutral\u002Fnone",[379,380,38,246,381,38],"45340523",27,"dynameds",[],[384,389,393,398,402,406],{"id":385,"name":386,"points":387,"category":388},"45231378","AnyCrap",1184,"creative",{"id":390,"name":391,"points":392,"category":388},"45220069","Beat Maker",202,{"id":394,"name":395,"points":396,"category":397},"45212535","Asxiv.org",161,"ai",{"id":399,"name":400,"points":401,"category":388},"45232299","CLAVIER-36",146,{"id":403,"name":404,"points":405,"category":388},"45247396","Backdrop",82,{"id":407,"name":408,"points":409,"category":410},"45245138","Dagger.js",80,"languages",false,"v2",{"id":414,"name":5,"line":6,"ai":411,"cat":20,"year":13,"points":14,"comments":15,"rec":28,"s":415,"top":18,"kw":416,"maker":11,"domain":10,"tier":19,"attempts":38,"strip":417,"url":9,"mix":418,"praise":74,"objection":94,"sub":21},45265831,-0.28,[24,25,26,27],"1131300551310010133310313313501011133353",[31,32,33,34],1793991130178]