Blog

AI insights, practical guides, and industry perspectives

When AI Says "It Can't Be Done as Is": Breaking Down Why and Finding Something Else to Measure (1)

When an AI or data specialist says something can't be done as is, the project usually stops right there. But that answer can be the starting point for redefining the problem rather than the end. Drawing on OpenAI's discontinued AI text classifier, Seoul's late-night bus routes, welfare checks based on electricity data and satellite-based poverty maps, this piece shows how to split the reasons into data, rules and accuracy, and how to use proxy metrics to measure what you can't measure directly.

AXAI transformationProblem definitionProxy metrics

Splitting Content Production Across AI — Weighing Sources, Rights and AI Disclosure Separately (2)

When AI finds your material, citing sources gets easier, but citing a source does not make the material yours to use. This article separates the questions: how to read KOGL and Creative Commons terms, AI voices and Korea's Unfair Competition Prevention Act, the US voice actors' lawsuit against Lovo, copyright registration standards for AI output, and the AI Framework Act's transparency guidelines alongside YouTube's rules on disclosing synthetic content.

AXAI transformationCopyrightAI disclosure

Splitting Content Production Across AI — How to Break Up the Process and Place Review Checkpoints (1)

With AI handling the script, the voice, the footage and the captions, one person can now produce a whole video. What remains is review. Drawing on how a broadcaster runs its AI anchor, the spread of low-quality AI video and YouTube's monetization policy, CNET's AI article errors, and the AP's generative AI standards, this article lays out how to divide content production into stages and where to put human checks between them.

AXAI transformationContent automationReview

A Digital Twin Sized for Small Businesses — What to Check Before Trusting a Simulator AI Built (2)

A simulator written by AI usually runs, and its graphs usually look plausible. But running and being right are not the same. Drawing on Google researchers' FEABench, the sinking of Norway's Sleipner A platform, the Columbia accident investigation and FDA guidance on simulation credibility, this article sets out four principles and a step-by-step plan for checking simulation results before using them in decisions: field yardsticks, sanity checks, labeling calibrated values and listing what the model cannot see.

AXAI TransformationDigital TwinAI Verification

A Digital Twin Sized for Small Businesses — Ask a Simulator First When a Real Test Is Expensive (1)

Decisions like changing a line, bringing in equipment or remaking a prototype cost a lot just to try once. Questions like these can be run through a small simulator first, and now that AI can write the analysis code, the barrier is even lower. Part 1 draws on Kia's Gwangmyeong plant, a Korean government smart factory program, the KISTI Virtual Design Center and small-business cases to set out which questions to give a simulator and which to avoid.

AXAI TransformationDigital TwinDecision Making

If AI Tools Are Being Copied Across Every Department: Build the Frame, Not Another Tool

Building a business tool with AI now takes a day. The trouble starts afterwards: every department makes its own copy, and soon nobody knows which version is current. From an AX perspective, this guide covers a design that separates shared parts from variable parts, the principle of leaving calculation to systems and narration to AI, and the questions to ask before you scale a tool out.

AXAI transformationCompany-wide rolloutGovernance

Test First with Fake Data Whose Answers You Know: Measuring Whether Fakes Resemble the Real Thing (2)

Even when you test an AI with synthetic data that has planted answers, scores get inflated if the fakes do not resemble the real thing. Part 2 draws on the inverse crime, domain randomization, the ISO image analysis standard and the Korean privacy regulator's synthetic data guide to show how to compare fakes and real data with the same yardstick, how to vary one condition at a time to find error sources, and an execution sequence that ends with real-world validation.

AXAI transformationValidationSynthetic data

Test First with Fake Data Whose Answers You Know: Building a Test with Planted Answers (1)

Validating an AI or automated measurement system requires correct answers, but field data often has none, and answers assigned by people are inconsistent. In that case you can first build a test with the answers planted in advance. Part 1 uses cases from Korea and abroad, including wine judging, mystery shopping, mutation testing and synthetic cell images, to show how to plant answers and how to build a test for each kind of task.

AXAI transformationValidationSynthetic data

Handing a Veteran's Intuition to AI — Exposing the Weights and Making Experts the Reviewers (2)

Once intuition has been broken into conditions, you still have to decide how to combine them, how far intuition can be trusted and where to place the experts. Drawing on Grove and colleagues' meta-analysis of clinical versus mechanical prediction, Kahneman and Klein's paper on the conditions for intuition, the dispute over expert compensation in a Korean government tacit-knowledge program and the WHO surgical safety checklist study, this article sets out the principles for finishing the criteria and an eight-step plan.

AXAI TransformationTacit KnowledgeManufacturing AI

Handing a Veteran's Intuition to AI — Breaking Intuition into Measurable Conditions (1)

An expert's intuition leaves the company on the day that person retires. What you can hand to AI is not the intuition itself but a set of criteria: the intuition broken into conditions, each with a way to measure it. Part 1 draws on Matsushita's bread maker, the Apgar score, a Korean government pilot program on tacit manufacturing knowledge and a press-line project to show how to split a veteran's judgment into named conditions and attach a measurement method to each.

AXAI TransformationTacit KnowledgeManufacturing AI

Make Information Gathering Your First AI Project: Automating an In-House Morning Briefing

If you are unsure where to start with AI, the information gathering someone on your team does by hand every day makes an excellent first project. From an AX perspective, this guide covers what to decide when automating a morning briefing of industry trends, regulatory notices and competitor news, how each department can use it, how to keep collection from silently stopping, and the internal rules that protect copyright and accuracy.

AXAI transformationWorkflow automationFirst project

Vision Inspection AI: Have You Decided What You Are Measuring? Resolution Limits and Measurement Variation (2)

A vision inspection AI cannot find anything smaller than what its camera can see, and if human judgments are inconsistent, it learns that inconsistency too. Part 2 covers working back from the smallest defect to the resolution you need, checking the measurement system first with Gage R&R, evaluating parts near the pass/fail line separately, and a pre-adoption checklist, with cases from Korea and abroad.

AXAI transformationVision inspectionQuality control

Vision Inspection AI: Have You Decided What You Are Measuring? Defining Size and Choosing a Reference Instrument (1)

More companies now ask about using cameras and AI to measure dimensions or catch defects. Yet the complaint we hear most right after rollout is that the AI does not match existing measurements. Usually the AI is not wrong; the project started without deciding what to measure and how. Part 1 covers writing down the definition of size and tying the reference instrument and units together, with cases from Korea and abroad.

AXAI transformationVision inspectionQuality control

Building an AI Product? Read the Failures First: Designing with Fix Lists and a Premortem (2)

Cause statements drawn from failure cases only become useful once they move into a design table and a meeting. This piece covers how to turn the change list of Lee Luda 2.0, a Korean chatbot that was shut down and then relaunched after fixes, into a checklist; how to run the premortem Gary Klein introduced in Harvard Business Review; and a six-step sequence for finishing a failure analysis in half a day.

AXAI transformationPremortemProduct planning

Building an AI Product? Read the Failures First: Extracting Cause Statements from Companies That Shut Down (1)

When someone proposes a new AI product or service, the first instinct is to look for success stories. Yet the information you need for design is more often left behind by companies that tried first and gave up. Using the five root causes of AI project failure identified by RAND, plus IBM Watson for Oncology, Zillow Offers, the Humane AI Pin and Korea's AI digital textbooks, this piece shows how to turn each failure into a one-sentence cause.

AXAI transformationFailure analysisProduct planning

Break Your AI Adoption Roadmap into Small Pieces: Why Big Projects Fail More Often

Companies that launch AI transformation as a single one-year mega-project are the ones that stall most often. Drawing on data showing a 61% success rate for small projects versus 6% for very large ones, this guide explains how to split an AI adoption roadmap into three layers (direction, quarterly goals and 2–4 week tasks), how to choose which tasks to start with, and the traps to avoid when you break the work down.

AXAI transformationAdoption strategyRoadmap

Pitfalls and a 4-Week Adoption Roadmap — The Record-Keeping AI Workflow (6)

We lay out the pitfalls you're bound to meet running the record-keeping AI workflow, and an adoption order split across four weeks so you don't burn out trying to do it all at once. Alongside AI coding tool adoption results from Korea and abroad, we propose the one thing to start today. The final article of the record-keeping AI workflow series.

AXAI transformationAI collaborationproductivityorganizational culture

Promoted to the Team's Language — Trackers, Wikis, and Git (5)

Personal memos and notes are read only by you. Decisions and procedures the team needs to know should be promoted to the issue tracker, the wiki, and the repository. We look at Toss's automatic documentation, architecture decision records, the GitLab handbook, the documentation cultures of Google and Spotify, and the DORA research on the link between documentation quality and performance. Part 5 of the record-keeping AI workflow series.

AXAI transformationcollaborationdocumentationknowledge management

Rules in One Place — How to Write an AI Rules File (2)

If you keep telling AI the same instructions over and over, it's time to gather those rules into one file. We lay out how rules files that load automatically at the start of a session work — CLAUDE.md, AGENTS.md, Cursor rules, GitHub Copilot repository instructions — and what separates a rule that gets followed from one that gets ignored. Part 2 of the record-keeping AI workflow series.

AXAI transformationAI collaborationdeveloper productivityprompting

Why AI Forgets Yesterday — Session Amnesia and the Cost of Context (1)

When a session ends, AI forgets everything it learned inside it. So you start over with the background every time, and re-investigate this week the cause you already tracked down last week. We look at the structural reason AI forgets yesterday, and the context cost that amnesia creates, backed by data from Korea and abroad. Part 1 of the record-keeping AI workflow series.

AXAI transformationAI collaborationproductivityknowledge management