
  <rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
    <channel>
      <title>Chaos and Order</title>
      <link>https://www.youngju.dev/blog</link>
      <description>천천히 올바르게. AI Researcher &amp; DevOps Engineer Youngju&#39;s tech blog. GPU/CUDA, LLM, MLOps, Kubernetes AI workloads, distributed training, and data engineering.</description>
      <language>ko</language>
      <managingEditor>fjvbn2003@gmail.com (Youngju Kim)</managingEditor>
      <webMaster>fjvbn2003@gmail.com (Youngju Kim)</webMaster>
      <lastBuildDate>Mon, 25 May 2026 00:00:00 GMT</lastBuildDate>
      <atom:link href="https://www.youngju.dev/tags/fico/feed.xml" rel="self" type="application/rss+xml"/>
      
  <item>
    <guid>https://www.youngju.dev/blog/culture/2026-05-25-credit-scoring-ai-underwriting-fico-vantagescore-kcb-nice-cic-upstart-affirm-klarna-2026-deep-dive.en</guid>
    <title>Credit Scoring and AI Underwriting in 2026 — FICO, VantageScore, KCB, NICE, CIC, Upstart, Affirm, Klarna Deep Dive</title>
    <link>https://www.youngju.dev/blog/culture/2026-05-25-credit-scoring-ai-underwriting-fico-vantagescore-kcb-nice-cic-upstart-affirm-klarna-2026-deep-dive.en</link>
    <description>In 2026, credit scoring is no longer an industry of paper applications and thick credit reports. The US FICO Score (`&lt;300/850 FICO`) and VantageScore have set the standard for nearly 50 years, while FICO XD, Experian Boost, and the FICO Resilience Index shake up the thin-file market. Upstart has crossed `$30B+` in cumulative originations using ML underwriting, while Affirm, Klarna, AfterPay, and Zilch have grown global BNPL into a `$300B BNPL market`. Korea runs a dual scale through KCB (Korea Credit Bureau) and NICE Credit (1-1000 points and 1-10 grades), and KakaoBank and Toss Bank deliver paperless loans with their own ML scoring. Japan operates 30-year credit databases through CIC, JICC, and KSC, supporting BNPL flows like AfterPay Japan and KakaoPay 후불결제. This article walks the entire map: the traditional 5-factor model (payment history, credit utilization, length, mix, new credit), the GBM and neural underwriting stack, alternative data and thin-file solutions, the Korean KCB/NICE scales, the integration of the three Japanese bureaus, BNPL regulation, and the rules of CFPB Reg B and the Korean FSS ADM guideline.</description>
    <pubDate>Mon, 25 May 2026 00:00:00 GMT</pubDate>
    <author>fjvbn2003@gmail.com (Youngju Kim)</author>
    <category>credit-scoring</category><category>ai-underwriting</category><category>fico</category><category>vantagescore</category><category>kcb-korea</category><category>nice-credit</category><category>cic-japan</category><category>jicc</category><category>upstart</category><category>affirm</category><category>klarna</category><category>afterpay</category><category>kakaopay-bnpl</category><category>toss-card</category><category>bnpl</category><category>cfpb-reg-b</category><category>fcra</category>
  </item>

  <item>
    <guid>https://www.youngju.dev/blog/culture/2026-05-25-credit-scoring-ai-underwriting-fico-vantagescore-kcb-nice-cic-upstart-affirm-klarna-2026-deep-dive.ja</guid>
    <title>信用スコアリングとAIアンダーライティング 2026 — FICO・VantageScore・KCB・NICE・CIC・Upstart・Affirm・Klarna 深掘りガイド</title>
    <link>https://www.youngju.dev/blog/culture/2026-05-25-credit-scoring-ai-underwriting-fico-vantagescore-kcb-nice-cic-upstart-affirm-klarna-2026-deep-dive.ja</link>
    <description>2026年の信用スコアリングは、もはや紙の申込書と分厚い信用報告書の産業ではない。米国のFICO Score(`&lt;300/850 FICO`)とVantageScoreが約50年にわたり標準を握る一方、FICO XD・Experian Boost・FICO Resilience Indexがthin file市場を揺さぶる。UpstartはMLアンダーライティングを武器に累計貸出 `$30B+` を突破し、Affirm・Klarna・AfterPay・Zilchがグローバルなbuy-now-pay-later市場を `$300B BNPL market` まで成長させた。韓国はKCB(コリアクレジットビューロー)とNICE信用評価が1-1000点・1-10等級の二重スケールを運営し、KakaoBank・Toss Bankが独自MLスコアで無書類融資を実現する。日本はCIC・JICC・KSCが分野別に30年保存の信用情報データベースを運営し、AfterPay日本やKakaoPay 後払いといったBNPLの流れを支えている。本稿は伝統的な5要因モデル(payment history・credit utilization・length・mix・new credit)から、GBM/Neural混在のAIスタック、alternative dataとthin file対応、韓国のKCB/NICEスコア体系、日本の3信用情報機関のデータ統合、BNPL規制、そしてCFPB Reg Bと韓国金融監督院ADMガイドラインまでを一気通貫で整理する。</description>
    <pubDate>Mon, 25 May 2026 00:00:00 GMT</pubDate>
    <author>fjvbn2003@gmail.com (Youngju Kim)</author>
    <category>credit-scoring</category><category>ai-underwriting</category><category>fico</category><category>vantagescore</category><category>kcb-korea</category><category>nice-credit</category><category>cic-japan</category><category>jicc</category><category>upstart</category><category>affirm</category><category>klarna</category><category>afterpay</category><category>kakaopay-bnpl</category><category>toss-card</category><category>bnpl</category><category>cfpb-reg-b</category><category>fcra</category>
  </item>

  <item>
    <guid>https://www.youngju.dev/blog/culture/2026-05-25-credit-scoring-ai-underwriting-fico-vantagescore-kcb-nice-cic-upstart-affirm-klarna-2026-deep-dive</guid>
    <title>신용평가와 AI Underwriting 2026 — FICO·VantageScore·KCB·NICE·CIC·Upstart·Affirm·Klarna 심층 가이드</title>
    <link>https://www.youngju.dev/blog/culture/2026-05-25-credit-scoring-ai-underwriting-fico-vantagescore-kcb-nice-cic-upstart-affirm-klarna-2026-deep-dive</link>
    <description>2026년의 신용평가는 더 이상 종이 신청서와 두꺼운 신용보고서의 산업이 아니다. 미국의 FICO Score(`&lt;300/850 FICO`)와 VantageScore가 50년 가까이 표준을 잡아온 한편, FICO XD·Experian Boost·FICO Resilience Index가 thin file 시장을 흔든다. Upstart는 ML 인수를 무기로 누적 대출 `$30B+`를 돌파했고, Affirm·Klarna·AfterPay·Zilch가 글로벌 BNPL 시장을 `$300B BNPL market`까지 키웠다. 한국은 KCB(코리아크레디트뷰로)와 NICE 신용평가가 1-1000점/1-10등급의 이중 척도를 운영하고, 카카오뱅크·토스뱅크가 자체 ML 신용평가로 무서류 대출을 만든다. 일본은 CIC·JICC·KSC가 30년 보존 신용정보 데이터베이스를 운영하며 AfterPay 재팬, KakaoPay 후불결제 같은 BNPL 흐름을 받쳐 준다. 이 글은 전통 5요인 모델(payment history·credit utilization·length·mix·new credit)에서 출발해, AI underwriting의 GBM/Neural 모델 스택, alternative data와 thin file 솔루션, 한국 KCB/NICE의 점수 체계, 일본 신용정보기관 3사의 데이터 통합, BNPL 규제, 그리고 CFPB Reg B와 한국 금감원 ADM 가이드라인까지 한 호흡으로 정리한다.</description>
    <pubDate>Mon, 25 May 2026 00:00:00 GMT</pubDate>
    <author>fjvbn2003@gmail.com (Youngju Kim)</author>
    <category>credit-scoring</category><category>ai-underwriting</category><category>fico</category><category>vantagescore</category><category>kcb-korea</category><category>nice-credit</category><category>cic-japan</category><category>jicc</category><category>upstart</category><category>affirm</category><category>klarna</category><category>afterpay</category><category>kakaopay-bnpl</category><category>toss-card</category><category>bnpl</category><category>cfpb-reg-b</category><category>fcra</category>
  </item>

    </channel>
  </rss>
