Donghyun Lee’s Trustworthy AI Noteshttps://donghyunlee.kr/en/Recent content on Donghyun Lee’s Trustworthy AI NotesHugo -- gohugo.ioen-USdonghyunlee.ai@gmail.com (Donghyun Lee)donghyunlee.ai@gmail.com (Donghyun Lee)© 2026 Donghyun LeeSat, 29 Aug 2026 00:00:00 +0000AI Has Arrived in Algal Bloom Forecasting — How Should We Read ‘Caution,’ ‘Warning,’ and a Seven-Day Forecast?https://donghyunlee.kr/en/posts/2026-08-29-ai-algal-bloom-forecast-literacy/Sat, 29 Aug 2026 00:00:00 +0000donghyunlee.ai@gmail.com (Donghyun Lee)https://donghyunlee.kr/en/posts/2026-08-29-ai-algal-bloom-forecast-literacy/Algal bloom alerts describe the observed present; AI and numerical models estimate change over the next seven days. This article explains how to read the forecast target, site, reference time, horizon, and uncertainty together.The AI-Written Python Code Ran—and It Was Wrong — Three Places Silent Errors Hidehttps://donghyunlee.kr/en/posts/2026-08-25-ai-code-quiet-errors/Fri, 21 Aug 2026 00:00:00 +0000donghyunlee.ai@gmail.com (Donghyun Lee)https://donghyunlee.kr/en/posts/2026-08-25-ai-code-quiet-errors/Code that finishes without an error is not verified code. This article explains three recurring hiding places for silent errors in AI-written Python—data-type assumptions, range boundaries, and swallowed exceptions—and a 30-second verification routine.Bookshttps://donghyunlee.kr/en/books/Sun, 16 Aug 2026 00:00:00 +0000donghyunlee.ai@gmail.com (Donghyun Lee)https://donghyunlee.kr/en/books/AI ERA SERIES — Korean-language books about what and how to learn in the age of AI.Does AI Really Say ‘I Don’t Know’ When It Encounters Unfamiliar Data? — Dataset Shift and OODhttps://donghyunlee.kr/en/posts/2026-08-16-dataset-shift-ood/Sun, 16 Aug 2026 00:00:00 +0000donghyunlee.ai@gmail.com (Donghyun Lee)https://donghyunlee.kr/en/posts/2026-08-16-dataset-shift-ood/AI uncertainty does not automatically increase when a model encounters data unlike its training data. This article distinguishes OOD from three kinds of distribution shift and sets out criteria for stress testing and revalidation.When a New Infectious Disease Has Little Data, Which Country Should AI Learn From? — Infectious Disease Forecasting and Heterogeneous Transfer Learninghttps://donghyunlee.kr/en/posts/2026-08-11-heterogeneous-transfer-learning/Tue, 11 Aug 2026 00:00:00 +0000donghyunlee.ai@gmail.com (Donghyun Lee)https://donghyunlee.kr/en/posts/2026-08-11-heterogeneous-transfer-learning/In transfer learning for infectious-disease forecasting, models trained on data from the most dissimilar countries consistently outperformed those trained on the most similar countries. This article explains the 30-country experiment and the limitations that must accompany its results.Privacy Policyhttps://donghyunlee.kr/en/privacy/Mon, 27 Jul 2026 00:00:00 +0000donghyunlee.ai@gmail.com (Donghyun Lee)https://donghyunlee.kr/en/privacy/How donghyunlee.kr processes personal information and uses Google Analytics.How Does AI Calculate ‘I Don’t Know’? — MC Dropout vs. Deep Ensembleshttps://donghyunlee.kr/en/posts/2026-07-26-mc-dropout-vs-deep-ensembles/Sun, 26 Jul 2026 00:00:00 +0000donghyunlee.ai@gmail.com (Donghyun Lee)https://donghyunlee.kr/en/posts/2026-07-26-mc-dropout-vs-deep-ensembles/MC Dropout runs one model in many forms; deep ensembles train multiple models separately. Twenty accessible points explain how these two approaches calculate what AI does not know, along with their costs and limitations.Why Do We Trust Claims More Easily When They Include Numbers? — 20 Short Thoughts on Reading Numbershttps://donghyunlee.kr/en/posts/2026-07-25-why-we-trust-numbers/Sat, 25 Jul 2026 00:00:00 +0000donghyunlee.ai@gmail.com (Donghyun Lee)https://donghyunlee.kr/en/posts/2026-07-25-why-we-trust-numbers/Why does a claim with a decimal point feel more objective? Twenty short thoughts on recovering the definitions, denominators, comparisons, uncertainty, and uses hidden behind a number.What Will Change First in the AI Era of 2027? — How Work Will Change Faster Than Jobshttps://donghyunlee.kr/en/posts/2026-07-17-ai-era-2027/Fri, 17 Jul 2026 00:00:00 +0000donghyunlee.ai@gmail.com (Donghyun Lee)https://donghyunlee.kr/en/posts/2026-07-17-ai-era-2027/In 2027, what may matter more than whether we use AI is the ability to design goals, context, tools, and verification as one system—and to improve the rubrics used to evaluate the results.If AI Writes All the Code, Do We Still Need to Learn Python? — From Writing to Verificationhttps://donghyunlee.kr/en/posts/2026-07-14-python-still-needed/Tue, 14 Jul 2026 00:00:00 +0000donghyunlee.ai@gmail.com (Donghyun Lee)https://donghyunlee.kr/en/posts/2026-07-14-python-still-needed/In an age when AI can write all the code, the reason to learn Python lies not in writing code but in verifying it. When should we trust or question code that fails silently, or AI that invents missing data to produce an answer?Why AI Predictions Should Not Give a Single Number — From Point Estimates to Confidence Intervalshttps://donghyunlee.kr/en/posts/2026-07-07-ai-uncertainty-interval/Sun, 12 Jul 2026 00:00:00 +0000donghyunlee.ai@gmail.com (Donghyun Lee)https://donghyunlee.kr/en/posts/2026-07-07-ai-uncertainty-interval/Just as statistics moved from point estimates to confidence intervals, AI is entering an era in which it must report not just one answer, but how much that answer can be trusted. Part 1 of the Trustworthy AI series.Abouthttps://donghyunlee.kr/en/about/Sat, 11 Jul 2026 00:00:00 +0000donghyunlee.ai@gmail.com (Donghyun Lee)https://donghyunlee.kr/en/about/Donghyun Lee — professor at Hankuk University of Foreign Studies and founder of AI Korea Inc., applying trustworthy AI to high-stakes forecasting.