[{"data":1,"prerenderedAt":899},["ShallowReactive",2],{"blog:\u002Fblog\u002F2026-04-30-scam-alert-pie":3},{"id":4,"title":5,"body":6,"date":889,"description":49,"extension":890,"image":891,"meta":892,"navigation":893,"path":894,"seo":895,"stem":896,"summary":897,"__hash__":898},"blog\u002Fblog\u002F2026-04-30-scam-alert-pie.md","Scam Alert Pie: A UX Case Study Against Warning Blindness",{"type":7,"value":8,"toc":855},"minimark",[9,13,18,22,25,28,31,39,42,53,56,62,65,69,72,75,78,81,103,106,109,112,118,122,125,131,134,145,153,166,169,175,178,182,185,188,217,220,223,227,230,235,240,243,246,249,255,258,264,268,271,274,276,282,285,288,305,309,312,315,317,323,326,330,333,336,339,342,348,351,354,357,360,364,368,371,377,381,384,390,394,397,403,407,410,413,419,425,431,437,443,449,455,461,467,470,476,480,483,486,561,564,567,571,574,577,583,586,589,593,599,602,606,609,615,618,621,638,641,645,648,651,657,660,666,669,672,675,679,683,686,692,695,699,702,705,711,714,718,721,724,730,734,737,740,746,749,753,756,762,765,768,774,777,781,784,787,792,795,812,814,843,846,852],[10,11,5],"h1",{"id":12},"scam-alert-pie-a-ux-case-study-against-warning-blindness",[14,15,17],"h2",{"id":16},"turning-a-scam-warning-into-a-protective-behavior-loop","Turning a scam warning into a protective behavior loop",[19,20,21],"p",{},"A job-search product needs to warn users about scams.",[19,23,24],{},"That sounds simple until the warning becomes familiar.",[19,26,27],{},"The first time a user sees a red safety box, they may read it. The fifth time, they skim it. The twentieth time, the brain compresses it into a label: \"the scam warning.\" After that, it may stop being read at all.",[19,29,30],{},"This is the design failure: a warning can remain visible while becoming behaviorally dead.",[19,32,33,34,38],{},"The case study here is a small UX pattern for preventing that collapse. I call it ",[35,36,37],"strong",{},"Scam Alert Pie",".",[19,40,41],{},"The invariant is:",[43,44,50],"pre",{"className":45,"code":47,"language":48,"meta":49},[46],"language-text","Danger -> [Micro-lesson] -> Antidote\n","text","",[51,52,47],"code",{"__ignoreMap":49},[19,54,55],{},"Or in product terms:",[43,57,60],{"className":58,"code":59,"language":48,"meta":49},[46],"Static safety frame + occasional rotating atomic tip + fixed protective CTA\n",[51,61,59],{"__ignoreMap":49},[19,63,64],{},"The point is not to lecture the user. The point is to interrupt a risky hiring flow with a tiny repeatable action that works even when the user is tired, hopeful, and under pressure.",[14,66,68],{"id":67},"the-context","The Context",[19,70,71],{},"Hiring products sit inside an uncomfortable threat model.",[19,73,74],{},"Scammers can impersonate recruiters, post fake roles, push malicious repos, request \"project reviews,\" harvest documents, extract wallet actions, or rent a candidate's attention through a fake opportunity. The attack does not always begin with an obvious theft request. It often starts with a normal-looking hiring ritual.",[19,76,77],{},"The user is also rarely in an ideal state.",[19,79,80],{},"They may be:",[82,83,84,88,91,94,97,100],"ul",{},[85,86,87],"li",{},"tired after rejections;",[85,89,90],{},"financially pressured;",[85,92,93],{},"excited by a rare opportunity;",[85,95,96],{},"afraid of seeming paranoid;",[85,98,99],{},"unwilling to lose a possible offer;",[85,101,102],{},"too depleted to read a full safety guide.",[19,104,105],{},"That matters. Safety UX must not assume a calm expert user. It must work for a person who has low energy, low spare attention, and a strong emotional reason to continue.",[19,107,108],{},"So the job of the interface is not just to display information.",[19,110,111],{},"It must create a protective behavior loop:",[43,113,116],{"className":114,"code":115,"language":48,"meta":49},[46],"risk noticed -> safe action remembered -> user slows down\n",[51,117,115],{"__ignoreMap":49},[14,119,121],{"id":120},"why-static-warnings-decay","Why Static Warnings Decay",[19,123,124],{},"The ordinary design move is predictable:",[43,126,129],{"className":127,"code":128,"language":48,"meta":49},[46],"Show a red warning.\nSay \"Be careful, this might be a scam.\"\nLink to a long guide.\n",[51,130,128],{"__ignoreMap":49},[19,132,133],{},"This can work once. It does not necessarily keep working.",[19,135,136,137,144],{},"Security-warning research describes habituation as a declining response to repeated exposure. A longitudinal study by Vance, Jenkins, Anderson, Bjornn, and Kirwan found attention and adherence to security warnings dropped over time, while polymorphic warnings reduced that drop compared with standard warnings (",[138,139,143],"a",{"href":140,"rel":141},"https:\u002F\u002Fscholarsarchive.byu.edu\u002Ffacpub\u002F6495\u002F",[142],"nofollow","BYU ScholarsArchive",").",[19,146,147,148,144],{},"UX research around banner blindness points to a related failure mode: users learn to ignore elements that look like a familiar class of low-value UI, especially when those elements sit in expected places or resemble promotional material (",[138,149,152],{"href":150,"rel":151},"https:\u002F\u002Fwww.nngroup.com\u002Farticles\u002Fbanner-blindness-old-and-new-findings\u002F",[142],"Nielsen Norman Group",[19,154,155,156,160,161,144],{},"There is an additional danger in warning design. A SOUPS 2019 paper, ",[157,158,159],"em",{},"The Fog of Warnings",", describes how habituation can generalize from frequent non-security notifications to security warnings when they share a similar look and feel (",[138,162,165],{"href":163,"rel":164},"https:\u002F\u002Fwww.usenix.org\u002Fconference\u002Fsoups2019\u002Fpresentation\u002Fvance",[142],"USENIX",[19,167,168],{},"The practical takeaway is simple:",[43,170,173],{"className":171,"code":172,"language":48,"meta":49},[46],"Static warnings degrade from message into shape.\n",[51,174,172],{"__ignoreMap":49},[19,176,177],{},"The user no longer reads the content. They recognize the container and move on.",[14,179,181],{"id":180},"the-product-problem","The Product Problem",[19,183,184],{},"In a hiring product, this is especially dangerous because the user is not merely making a low-stakes UI choice.",[19,186,187],{},"They may be one click away from:",[82,189,190,193,196,199,202,205,208,211,214],{},[85,191,192],{},"opening a malicious file;",[85,194,195],{},"running a repo;",[85,197,198],{},"sending documents;",[85,200,201],{},"sharing personal contact data;",[85,203,204],{},"joining a suspicious call;",[85,206,207],{},"performing unpaid work;",[85,209,210],{},"logging in through a personal account;",[85,212,213],{},"connecting a wallet;",[85,215,216],{},"continuing a manipulative conversation.",[19,218,219],{},"The warning must do more than say \"danger.\"",[19,221,222],{},"It must help the user perform a safer next move with very low effort.",[14,224,226],{"id":225},"the-pattern-scam-alert-pie","The Pattern: Scam Alert Pie",[19,228,229],{},"The pattern has three slices.",[43,231,233],{"className":232,"code":47,"language":48,"meta":49},[46],[51,234,47],{"__ignoreMap":49},[236,237,239],"h3",{"id":238},"_1-danger","1. Danger",[19,241,242],{},"This is the stable safety frame.",[19,244,245],{},"It should remain recognizable. The user must immediately understand that they are in a risk zone.",[19,247,248],{},"Example:",[43,250,253],{"className":251,"code":252,"language":48,"meta":49},[46],"This contact may be unverified.\nDo not share sensitive materials before verification.\n",[51,254,252],{"__ignoreMap":49},[19,256,257],{},"The purpose is not to explain every scam pattern. It answers only:",[43,259,262],{"className":260,"code":261,"language":48,"meta":49},[46],"What is happening?\nWhy should I slow down?\n",[51,263,261],{"__ignoreMap":49},[236,265,267],{"id":266},"_2-micro-lesson","2. Micro-lesson",[19,269,270],{},"This is the optional rotating middle line.",[19,272,273],{},"It appears sometimes, not always. It carries one specific safety atom.",[19,275,248],{},[43,277,280],{"className":278,"code":279,"language":48,"meta":49},[46],"Tip: High salary is not proof. Company-domain contact is closer to proof.\n",[51,281,279],{"__ignoreMap":49},[19,283,284],{},"The purpose is to break semantic caching without turning the interface into noise.",[19,286,287],{},"The tip must be:",[82,289,290,293,296,299,302],{},[85,291,292],{},"short;",[85,294,295],{},"specific;",[85,297,298],{},"actionable;",[85,300,301],{},"non-abstract;",[85,303,304],{},"limited to one idea.",[236,306,308],{"id":307},"_3-antidote","3. Antidote",[19,310,311],{},"This is the fixed protective action.",[19,313,314],{},"It should be stable because the user needs to learn one repeatable move.",[19,316,248],{},[43,318,321],{"className":319,"code":320,"language":48,"meta":49},[46],"Not sure? Run Magic Scam Check.\n[Copy prompt]\n",[51,322,320],{"__ignoreMap":49},[19,324,325],{},"The antidote matters because anxiety alone is not a safety flow. If the product only increases fear, the user may freeze, dismiss the warning, or continue anyway. A good warning turns uncertainty into a concrete next action.",[14,327,329],{"id":328},"why-the-middle-changes-but-the-cta-does-not","Why the Middle Changes but the CTA Does Not",[19,331,332],{},"There are two bad extremes.",[19,334,335],{},"If everything is static, the alert becomes wallpaper.",[19,337,338],{},"If everything changes, the alert becomes noise.",[19,340,341],{},"The useful middle is:",[43,343,346],{"className":344,"code":345,"language":48,"meta":49},[46],"stable shell + variable signal + stable action\n",[51,347,345],{"__ignoreMap":49},[19,349,350],{},"The stable shell preserves recognition.",[19,352,353],{},"The variable signal restores attention.",[19,355,356],{},"The stable action trains one protective habit.",[19,358,359],{},"That is the core UX move.",[14,361,363],{"id":362},"example-ui-copy","Example UI Copy",[236,365,367],{"id":366},"compact-default","Compact Default",[19,369,370],{},"Use this for repeated exposure when no specific high-risk pattern has been detected.",[43,372,375],{"className":373,"code":374,"language":48,"meta":49},[46],"Unverified contact.\nDo not share sensitive materials before verification.\n\n[Run Magic Scam Check]\n",[51,376,374],{"__ignoreMap":49},[236,378,380],{"id":379},"expanded-with-tip","Expanded With Tip",[19,382,383],{},"Use this for first exposure, periodic reinforcement, or a relevant detected risk pattern.",[43,385,388],{"className":386,"code":387,"language":48,"meta":49},[46],"This contact may be unverified.\nDo not share sensitive materials before verification.\n\nTip: If they ask for a project review before verification, stop and verify first.\n\nNot sure? Run Magic Scam Check.\n[Copy prompt]\n",[51,389,387],{"__ignoreMap":49},[236,391,393],{"id":392},"high-risk-technical-action","High-Risk Technical Action",[19,395,396],{},"Use this before repo execution, external files, APKs, browser extensions, wallet actions, or account authorization.",[43,398,401],{"className":399,"code":400,"language":48,"meta":49},[46],"High-risk step.\nDo not run code, install software, connect a wallet, or log in through a personal account before verification.\n\nTip: A repo is not proof. It can be the attack surface.\n\n[Run Magic Scam Check]\n",[51,402,400],{"__ignoreMap":49},[14,404,406],{"id":405},"atomic-tip-library","Atomic Tip Library",[19,408,409],{},"The tip library is the product's small safety curriculum.",[19,411,412],{},"Examples:",[43,414,417],{"className":415,"code":416,"language":48,"meta":49},[46],"High salary is not proof. Verified company-domain contact is closer to proof.\n",[51,418,416],{"__ignoreMap":49},[43,420,423],{"className":421,"code":422,"language":48,"meta":49},[46],"If they ask for a project review before verification, stop and verify first.\n",[51,424,422],{"__ignoreMap":49},[43,426,429],{"className":427,"code":428,"language":48,"meta":49},[46],"A real recruiter should be able to prove who they represent.\n",[51,430,428],{"__ignoreMap":49},[43,432,435],{"className":433,"code":434,"language":48,"meta":49},[46],"Do not open files or repo links from unverified recruiters.\n",[51,436,434],{"__ignoreMap":49},[43,438,441],{"className":439,"code":440,"language":48,"meta":49},[46],"A vague role plus an urgent next step is a risk pattern.\n",[51,442,440],{"__ignoreMap":49},[43,444,447],{"className":445,"code":446,"language":48,"meta":49},[46],"If their cost to fake is near zero and your cost is high, slow down.\n",[51,448,446],{"__ignoreMap":49},[43,450,453],{"className":451,"code":452,"language":48,"meta":49},[46],"Before a call, ask for role scope, compensation range, company identity, and process.\n",[51,454,452],{"__ignoreMap":49},[43,456,459],{"className":457,"code":458,"language":48,"meta":49},[46],"Screenshots and documents can be forged. Company-domain contact is harder to fake.\n",[51,460,458],{"__ignoreMap":49},[43,462,465],{"className":463,"code":464,"language":48,"meta":49},[46],"Do not let a scammer rent your attention for free.\n",[51,466,464],{"__ignoreMap":49},[19,468,469],{},"The last one is especially useful because it reframes scam risk as adversarial economics:",[43,471,474],{"className":472,"code":473,"language":48,"meta":49},[46],"They pay almost nothing.\nYou pay with attention, hope, time, documents, access, and risk.\n",[51,475,473],{"__ignoreMap":49},[14,477,479],{"id":478},"trigger-logic","Trigger Logic",[19,481,482],{},"The rotating tip should not appear every time.",[19,484,485],{},"A reasonable starting policy:",[487,488,489,502],"table",{},[490,491,492],"thead",{},[493,494,495,499],"tr",{},[496,497,498],"th",{},"Situation",[496,500,501],{},"Alert Behavior",[503,504,505,514,522,530,538,545,553],"tbody",{},[493,506,507,511],{},[508,509,510],"td",{},"First exposure",[508,512,513],{},"Expanded alert + tip + CTA",[493,515,516,519],{},[508,517,518],{},"Normal repeated exposure",[508,520,521],{},"Compact alert + CTA",[493,523,524,527],{},[508,525,526],{},"Every 3-5 exposures",[508,528,529],{},"Alert + rotating tip + CTA",[493,531,532,535],{},[508,533,534],{},"High-risk pattern detected",[508,536,537],{},"Alert + relevant tip + CTA",[493,539,540,543],{},[508,541,542],{},"Before revealing contact",[508,544,521],{},[493,546,547,550],{},[508,548,549],{},"Before external link, file, or project review",[508,551,552],{},"Specific tip + CTA",[493,554,555,558],{},[508,556,557],{},"Before local code execution or wallet action",[508,559,560],{},"High-risk alert + specific tip + CTA",[19,562,563],{},"The principle is not randomness. It is controlled variation.",[19,565,566],{},"The product keeps the protective object recognizable while refreshing the part that the user is likely to stop reading.",[14,568,570],{"id":569},"magic-scam-check","Magic Scam Check",[19,572,573],{},"The antidote should not require the user to become a scam-detection expert.",[19,575,576],{},"The better move is:",[43,578,581],{"className":579,"code":580,"language":48,"meta":49},[46],"uncertain -> copy prompt -> paste conversation -> get verdict + next safe step\n",[51,582,580],{"__ignoreMap":49},[19,584,585],{},"Most job seekers already have access to ChatGPT, Gemini, Claude, or another LLM. The product can use that fact defensively. It does not need to build a perfect internal anti-scam system before helping the user.",[19,587,588],{},"It can give the user a ready-to-run safety protocol.",[236,590,592],{"id":591},"prompt-skeleton","Prompt Skeleton",[43,594,597],{"className":595,"code":596,"language":48,"meta":49},[46],"You are my hiring-scam risk analyst.\n\nAnalyze this job or recruiter interaction as an adversarial economic system.\n\nCore rule:\nA legitimate hiring process should have a roughly fair exchange of effort and verifiable artifacts.\nA suspicious process often asks me to spend high-value resources: time, attention, documents, code, calls, wallet actions, KYC, project review, or account access, while the other side provides low-cost or easily faked artifacts.\n\nEvaluate:\n1. What are they asking from me?\n2. What have they proven?\n3. What is their cost to fake this interaction?\n4. What is my cost or risk if I continue?\n5. Are there asymmetric effort, urgency, vagueness, unverifiable identity, fake authority, or project-review bait patterns?\n6. What minimum verification should I request before continuing?\n7. Give a verdict: Green \u002F Yellow \u002F Red.\n8. Give one safe next message I can send.\n\nImportant:\nDo not be polite to the opportunity.\nProtect my time, documents, devices, wallets, credentials, and attention.\nIf information is missing, tell me exactly what to ask for.\n",[51,598,596],{"__ignoreMap":49},[19,600,601],{},"This prompt is intentionally strict. A low-resource user does not need a polite opportunity coach. They need a second mind that will protect them from asymmetric harm.",[14,603,605],{"id":604},"reporting-loop","Reporting Loop",[19,607,608],{},"The next layer is to turn individual protection into platform intelligence.",[43,610,613],{"className":611,"code":612,"language":48,"meta":49},[46],"user reports suspicious contact\n-> report becomes structured evidence\n-> platform cross-checks patterns\n-> trust score changes\n-> confirmed patterns become new atomic tips\n",[51,614,612],{"__ignoreMap":49},[19,616,617],{},"This can improve the product over time, but it needs an abuse-resistant design.",[19,619,620],{},"If rewards are added, reward structure should be conservative:",[82,622,623,626,629,632,635],{},[85,624,625],{},"no reward for raw accusation;",[85,627,628],{},"reward only after cross-check;",[85,630,631],{},"higher reward for reusable evidence;",[85,633,634],{},"no reward for noisy reports;",[85,636,637],{},"trusted reporter score over time.",[19,639,640],{},"The goal is not to create accusation spam. The goal is to convert user observations into safety intelligence.",[14,642,644],{"id":643},"a-second-order-signal","A Second-Order Signal",[19,646,647],{},"There is a possible second-order effect worth watching.",[19,649,650],{},"Ordinary warnings often end silently:",[43,652,655],{"className":653,"code":654,"language":48,"meta":49},[46],"warning shown -> user avoided risk -> event forgotten\n",[51,656,654],{"__ignoreMap":49},[19,658,659],{},"Scam Alert Pie may create a more shareable user story:",[43,661,664],{"className":662,"code":663,"language":48,"meta":49},[46],"risk noticed -> check ran -> proof requested -> interaction clarified -> lesson remembered\n",[51,665,663],{"__ignoreMap":49},[19,667,668],{},"The user does not only avoid a bad interaction. They get a small narrative: I noticed the asymmetry, ran the check, asked for proof, and watched the situation reveal itself.",[19,670,671],{},"For product teams, this matters because success may not show up only as fewer victims. It may also show up as users sharing screenshots, reporting suspicious flows, reusing the verification challenge elsewhere, and describing the moment they realized they were about to spend expensive trust against cheap proof.",[19,673,674],{},"This should be treated as a product hypothesis, not a promised therapeutic outcome. But it is a useful thing to instrument and listen for.",[14,676,678],{"id":677},"what-this-case-study-demonstrates","What This Case Study Demonstrates",[236,680,682],{"id":681},"adversarial-economics","Adversarial Economics",[19,684,685],{},"The scam is modeled as a cost imbalance:",[43,687,690],{"className":688,"code":689,"language":48,"meta":49},[46],"attacker cost to fake ~= low\nuser cost to process ~= high\n",[51,691,689],{"__ignoreMap":49},[19,693,694],{},"If the other side cheaply generates the appearance of opportunity while the user pays with attention, time, documents, calls, code execution, or project review, the product should interrupt the flow.",[236,696,698],{"id":697},"low-resource-ux","Low-Resource UX",[19,700,701],{},"The user is not modeled as an ideal rational analyst.",[19,703,704],{},"The user is modeled as:",[43,706,709],{"className":707,"code":708,"language":48,"meta":49},[46],"tired\nhopeful\nstressed\nfinancially pressured\nattention-limited\nafraid to lose the opportunity\n",[51,710,708],{"__ignoreMap":49},[19,712,713],{},"That assumption changes the design. It produces a short protective move instead of a long educational demand.",[236,715,717],{"id":716},"attention-aware-safety-design","Attention-Aware Safety Design",[19,719,720],{},"Static warnings are not enough because attention adapts.",[19,722,723],{},"Scam Alert Pie uses stability where recognition matters and controlled variation where attention decays.",[43,725,728],{"className":726,"code":727,"language":48,"meta":49},[46],"Stable shell.\nRotating atom.\nFixed action.\n",[51,729,727],{"__ignoreMap":49},[236,731,733],{"id":732},"ai-as-protective-companion","AI as Protective Companion",[19,735,736],{},"The LLM is not used to generate more hiring spam.",[19,738,739],{},"It is used as a defensive layer:",[43,741,744],{"className":742,"code":743,"language":48,"meta":49},[46],"parse the situation\ndetect asymmetric extraction\nask for missing verification\ngenerate safe next message\n",[51,745,743],{"__ignoreMap":49},[19,747,748],{},"This is an important product direction. AI can amplify attackers, but it can also give vulnerable users a protective analyst at the exact moment when they need one.",[236,750,752],{"id":751},"turning-incident-into-mechanism","Turning Incident Into Mechanism",[19,754,755],{},"The case begins with a messy real-world event:",[43,757,760],{"className":758,"code":759,"language":48,"meta":49},[46],"suspicious recruiter \u002F risky hiring interaction\n",[51,761,759],{"__ignoreMap":49},[19,763,764],{},"But the output is not just a complaint.",[19,766,767],{},"It becomes:",[43,769,772],{"className":770,"code":771,"language":48,"meta":49},[46],"criterion\nprompt\nUX pattern\nreporting loop\ngeneralizable safety mechanism\n",[51,773,771],{"__ignoreMap":49},[19,775,776],{},"That is the builder move: convert a live failure mode into productizable protection.",[14,778,780],{"id":779},"why-this-generalizes","Why This Generalizes",[19,782,783],{},"The invariant is not hiring.",[19,785,786],{},"The invariant is the pie:",[43,788,790],{"className":789,"code":47,"language":48,"meta":49},[46],[51,791,47],{"__ignoreMap":49},[19,793,794],{},"It applies anywhere there is:",[82,796,797,800,803,806,809],{},[85,798,799],{},"repeated risk;",[85,801,802],{},"a tired user;",[85,804,805],{},"information asymmetry;",[85,807,808],{},"low attacker cost;",[85,810,811],{},"high user cost of error.",[19,813,412],{},[82,815,816,819,822,825,828,831,834,837,840],{},[85,817,818],{},"job scams;",[85,820,821],{},"crypto wallet warnings;",[85,823,824],{},"phishing links;",[85,826,827],{},"marketplace scams;",[85,829,830],{},"freelance project bait;",[85,832,833],{},"fake investment offers;",[85,835,836],{},"malware repo links;",[85,838,839],{},"dating and social-engineering scams;",[85,841,842],{},"KYC or document harvesting.",[19,844,845],{},"The final rule is simple:",[43,847,850],{"className":848,"code":849,"language":48,"meta":49},[46],"Do not lecture.\nDo not only frighten.\nDo not overload.\n\nName the danger.\nSometimes teach one small lesson.\nAlways give one protective action.\n",[51,851,849],{"__ignoreMap":49},[19,853,854],{},"That is Scam Alert Pie.",{"title":49,"searchDepth":856,"depth":856,"links":857},2,[858,859,860,861,862,868,869,874,875,876,879,880,881,888],{"id":16,"depth":856,"text":17},{"id":67,"depth":856,"text":68},{"id":120,"depth":856,"text":121},{"id":180,"depth":856,"text":181},{"id":225,"depth":856,"text":226,"children":863},[864,866,867],{"id":238,"depth":865,"text":239},3,{"id":266,"depth":865,"text":267},{"id":307,"depth":865,"text":308},{"id":328,"depth":856,"text":329},{"id":362,"depth":856,"text":363,"children":870},[871,872,873],{"id":366,"depth":865,"text":367},{"id":379,"depth":865,"text":380},{"id":392,"depth":865,"text":393},{"id":405,"depth":856,"text":406},{"id":478,"depth":856,"text":479},{"id":569,"depth":856,"text":570,"children":877},[878],{"id":591,"depth":865,"text":592},{"id":604,"depth":856,"text":605},{"id":643,"depth":856,"text":644},{"id":677,"depth":856,"text":678,"children":882},[883,884,885,886,887],{"id":681,"depth":865,"text":682},{"id":697,"depth":865,"text":698},{"id":716,"depth":865,"text":717},{"id":732,"depth":865,"text":733},{"id":751,"depth":865,"text":752},{"id":779,"depth":856,"text":780},"2026-04-30","md",null,{},true,"\u002Fblog\u002F2026-04-30-scam-alert-pie",{"title":5,"description":49},"blog\u002F2026-04-30-scam-alert-pie","How a static scam warning can become wallpaper, and how a small Danger -> Micro-lesson -> Antidote pattern can protect low-energy users in hiring products.","6V4aiMsIqR1hek6BjQ-Vu6Y66pjxgHU5mrdXhS7KNfA",1777518596464]