Table of Contents
Artificial Intelligence Translates the Language of Animals
Kim Kyung-jin, Attorney at Law
The Story of AI Learning to Listen to Whales, Dolphins, Birds, and Bees
Twelve chapters on how AI listens to dolphins, sperm whales, humpback whales, birds, and bees to find rules in their sounds, and what this technology means for its risks and for animal rights. Written in simple sentences a child can read, with verified sources in every section.
Table of Contents
AI Deciphers Ancient Scripts
Kim Kyung-jin, Attorney at Law
Ancient Records Revived by AI
In twelve chapters, this book explains how AI revives records once unreadable, from burned scrolls and wooden slips buried in mud to broken clay tablets. It covers virtual unrolling at Herculaneum, virtual collation of oracle-bone texts, reading Silla wooden tablets, computational analysis of undeciphered scripts, and multispectral archives, with verified references for each chapter.
Table of Contents
Artificial Intelligence for New Materials Design and Rocket Propulsion Engineering
Kim Kyung-jin, Attorney at Law
AI Potentials, Self-Driving Laboratories, and Physics-Informed Machine Learning (PIML)
Ten chapters on how artificial intelligence is changing new materials and rocket propulsion: atomic simulation, generative models, self-driving labs, high-temperature alloys, metal 3D printing, combustion, cooling design, and engine diagnosis and control. Written without equations, with verified sources in every chapter.
AI Library
The Age of Autonomous Scientific Discovery
Kim Kyung-jin, Attorney at Law
AI Scientists and Self-Driving Labs
This book follows how AI scientists and self-driving labs are changing the way science generates and verifies claims. It covers literature-based discovery, natural-language protocols translated into robot commands, multi-agent research systems, closed-loop laboratories, materials search, the verification gap, chains of evidence, research harnesses, journal ethics, and legal responsibility.
AI Library
A New Era of Life Sciences Opened by Artificial Intelligence
Structural Proteomics, Genomic Foundation Models, Autonomous Laboratories, and Global Governance
Kim Kyung-jin, Attorney at Law
This book is a research volume compiled with artificial intelligence. A human selected the materials and structured the work, while AI models drafted the sentences and cross-checked the facts.
AI Library
The Double Structure of Digital Sovereignty
Europe’s Departure from Palantir and the Chains of American Big Tech
Kim Kyung-jin, Attorney at Law
This is a record of 2026, when European intelligence agencies and defense ministries began removing analytics tools from America’s Palantir. It covers the replacement decisions made by France’s General Directorate for Internal Security (DGSI), Germany’s Federal Office for the Protection of the Constitution (BfV), and the Netherlands Ministry of Defense; the incident in which US export controls severed an ally’s ac…
New English Edition
Artificial Intelligence in Horticulture
Kim Kyung-jin, Attorney at Law
Across five chapters and ten sections, this book examines computer vision for crop diagnosis, harvesting robots and autonomous field systems, smart greenhouses and digital twins, precision irrigation and supply-chain quality control, high-throughput phenotyping, and predictive breeding.
New English Edition
Artificial Intelligence in Food Crop Agriculture
Kim Kyung-jin, Attorney at Law
Across six chapters and eighteen sections, the book examines digital agricultural infrastructure, remote sensing, crop diagnosis, yield forecasting, precision irrigation, genomics, molecular breeding, agricultural robotics, climate-smart agriculture, and global food security.
New English Edition
The Future of Forestry and Agroforestry
Kim Kyung-jin, Attorney at Law
Driven by Artificial Intelligence and Digital Innovation
Across five chapters and fifteen sections, the book follows satellites, drones, LiDAR, digital twins, forest-specific language models, wildfire and pest forecasting, forestry robotics, agroforestry, timber traceability, and forest carbon markets.
New English Edition
Smart Livestock Farming: AI Enters the Barn
Kim Kyung-jin, Attorney at Law
Sensors listen, cameras watch, and artificial intelligence helps farmers decide.
Across five chapters and fifteen sections, the book follows precision livestock farming from animal health and reproduction to robotic milking, virtual fencing, digital twins, methane reduction, welfare, and data ownership.
Table of Contents
Han Dong-hoon, Busan Buk-gu Gap: A Record of the 100 Days Before and After the Election (Mar. 26-Jul. 3, 2026)
Kim Kyung-jin
Table of Contents and 13 sections
From March 26 to July 3, 2026, this record follows the spring after expulsion, the Busan Buk-gu Gap by-election, victory as an independent, and the first bill submitted in the National Assembly.
Table of Contents
Artificial Intelligence and Medicine
Kim Kyung-jin, Attorney at Law
AI in clinical care, hospitals, education, and research
AI in medical imaging, risk prediction, treatment planning, hospital operations, education, and research, with patient safety, privacy, and accountability.
[AI Library] Chapter 24 Habits Become Skill
Get 1,000% More from Your Computer with AI Agents
Chapter 24 Habits Become Skill
Kim Kyung-jin
— Record, verify, and prepare to revert
Introduction: I Definitely Fixed It Yesterday, but What and How Did I Fix Again?
Last week, your computer slowed down, so you turned to an AI agent. The agent tuned various settings, making the machine snappy again. But a week later, a different issue arose. You want to check whether the previous setting changes were responsible, but you cannot recall what was changed.
Have you ever had an experience like this?
An AI agent is a tool. No matter how extraordinary a tool is, poor user habits lead to poor outcomes. Conversely, adopting a few simple habits enables far safer and more productive work with the exact same tool.
In this chapter, we explore three habits for using AI agents safely over the long haul: recording, verifying, and preparing to revert. It is not difficult; think of it as putting the five promises from Chapter 4 into natural daily practice.
24.1 The First Habit: Record Everything
Why must you record?
When you visit a clinic, the doctor keeps a medical record: when you arrived, what symptoms you reported, and what medication was prescribed. When you return, the doctor reviews that record to make informed decisions. Without records, they would have to ask everything from scratch each visit.
Computer maintenance works identically. When you retain logs of what you instructed the AI agent to do and what changes it made, you can pinpoint root causes rapidly if issues arise later. Without records, you do not even know where to begin looking.
The core rule of logging is straightforward: record four elements—When, What, Why, and How it was changed.
What should you record?
There are three primary items to record:
First: The exact prompt you sent to the AI. Retaining your verbatim request eliminates the need to reinvent prompts when repeating the task later.
Second: The commands executed by the AI and details of changes made—such as which files were modified and which settings altered. Terminal agents like Claude Code print executed commands directly on screen, so you can easily copy and archive them.
Third: The system state before and after the operation. Numerical records such as boot time reduced from 45s to 22s or disk usage dropped from 89% to 62% provide the proof needed to verify genuine improvement later.
The easiest logging method: Have the AI do the logging too
While logging is essential, typing notes manually every time is tedious. The best approach is delegating the logging to the AI as well by adding a single line to the end of your prompt.
Create a script that automatically organizes the Downloads folder by extension. When finished, append today's date, changes made, and executed commands to the work-log.txt file.
→ After building the folder organization script, the AI logs its actions to the text file automatically. Opening this file later gives you an immediate overview of what was done and when.
Summarize the work just executed in the format below and append it to change-log.md.
Format: Date / Task Summary / Pre-state / Post-state / Rollback Procedure
→ The AI logs the completed task in your designated format. Because it packages date, summary, before-and-after metrics, and rollback steps together, reviewing it later is effortless.
This eliminates the burden of manual logging: the AI does the work, and the AI keeps the record. All you do is open the file occasionally to review it.
Creating your personal work log
Scattered records are hard to locate later. It is best to consolidate them neatly in a single file—your Work Log.
Store your work log in an easily accessible location. For example, create an `AI_Work_Logs` folder inside the `Documents` folder on Windows or Mac, and keep your logs there.
A typical work log entry looks like this:
Date: 2026-08-15
Task: Startup program optimization
Pre-state: Boot time 48s, 23 startup apps
Post-state: Boot time 21s, 9 startup apps
Prompt executed: "Find and disable startup programs that can safely be turned off, and compare before-and-after boot times."
Rollback Procedure: Double-click backup file startup-backup-0815.reg to restore original settings.
Formatting this manually at first might feel bothersome. But providing this template to the AI once and asking Please log in this format ensures the AI follows the exact same structure automatically from then on.
Going forward, whenever a task I assign completes, automatically append Date, Task Summary, Pre-state, Post-state, and Rollback Procedure to log-2026-08.md inside the AI_Work_Logs folder.
→ Entering this prompt once early in a session ensures the AI maintains logs for each subsequent task. After a month, a full record of your activities is gathered in one file.
We recommend reviewing your work log once a month. Spotting recurring patterns—like I had to fix network issues three times this month—signals that it is time to address root causes.
24.2 The Second Habit: Verify with Numbers
Confirming through Numbers rather than Feelings
It feels a bit faster. This statement is not an objective assessment. You cannot know whether it truly sped up or if it is just a placebo effect. Recall the first promise from Chapter 4: Measurement comes before feeling.
Verification means confirming genuine improvement with concrete numbers. It is simple: just tell the AI, Please compare before and after.
For instance, if you optimized boot performance, place pre-tuning and post-tuning boot times side by side. If you cleaned up disk space, compare free space before and after. This comparative measurement constitutes verification.
Verify the results of the disk cleanup just executed. Compare free disk space before and after, number of deleted files, and utilization percentage relative to total capacity in a table.
→ The AI summarizes pre- and post-cleanup figures in a table, giving you clear answers like Free disk space increased from 12 GB to 38 GB.
Changing only one variable at a time
The second principle of verification is changing only one thing at a time—the second promise from Chapter 4.
Suppose your computer booted slowly, so you trimmed startup programs, altered antivirus settings, and adjusted power options simultaneously. Boot speed improved. While the outcome is good, you have no way of knowing which change delivered the benefit. If similar trouble occurs later, you must repeat all three.
Changing one thing at a time reveals the precise impact of each action. Even if it takes slightly longer, it is far more accurate and safe in the long run.
I want to improve boot speed. Please apply startup program cleanup, fast startup configuration, and disk cleanup one at a time in sequence. Measure boot times after each step to compare how much benefit each change provides.
→ The AI applies each optimization method sequentially, measuring boot times after each step. Yielding data like Startup cleanup reduced 15s; Fast Startup reduced an additional 3s tells you precisely which fix was most effective.
Re-verifying one week later
Fixing an issue today is not the end of the story. Certain problems resurface over time: memory leaks can refill RAM after several days, and disk clutter can accumulate again after a month.
It is good to make a one-week checkup a habit. Review your work log and verify whether last week's fixes are still holding. You can also delegate this to the AI.
I optimized startup programs one week ago. Re-examine the current startup list and report whether any new entries have been added since then.
→ The AI inspects the active startup list and flags any newly registered programs added since the cleanup, letting you take action immediately if unwanted software slipped back in.
Last month's disk cleanup brought free space to 38 GB. Check current free space and estimate when capacity will become constrained again at the current rate of consumption.
→ The AI inspects current disk usage trends over the past month and projects: Cleanup will likely be needed again in about 3 months, letting you plan proactively.
24.3 The Third Habit: Prepare to Revert
Why is rollback readiness so critical?
Writing with a pencil without an eraser feels uneasy because mistakes cannot be undone. Changing computer settings is identical: having a rollback eraser gives you peace of mind to try new configurations. Without an eraser, a single mistake can cause major trouble.
Recall the third promise from Chapter 4: Preserve pre-change states and rollback methods. This is not merely a rule, but a mandatory safety habit when working with AI agents.
Entrust rollback preparation to the AI
The simplest approach is adding Ensure this can be reverted into your prompt.
Change network DNS settings to Google DNS (8.8.8.8). Back up current settings before changing, and provide the command to restore original settings.
→ The AI saves current DNS settings to a backup file before applying new settings, providing instructions: Run this command to revert if needed.
Save current state as a backup file before optimizing startup programs. Ensure I can restore everything in one click if problems arise later.
→ The AI archives the startup list into a backup file. If you discover a certain program was needed after all, running the backup file restores the previous configuration instantly.
Both prompts share a common pattern: they request backup before changing and instructions on how to revert at the same time. That is the core point.
Caution: Always pause before high-risk operations
AI agents generally execute whatever tasks they are assigned. Thus, building the habit of pausing before risky operations is crucial.
In the four situations below, always say: Wait, back up before executing.
Caution: Deleting files or formatting drives
Caution: Modifying the Windows Registry (core system configuration repository)
Caution: Modifying boot-related parameters
Caution: Touching core system files or hardware drivers
If remembering all four is hard, simply remember: Back up first for anything hard to undo.
Assess whether this task poses risks to the system first. If risky, explain backup procedures and execute only after I explicitly approve.
→ The AI evaluates operational risk first. If risky, it suggests backups: This operation modifies the Registry. Shall we create a Restore Point first? and waits until you say Yes.
This prompt is especially recommended because it prevents dangerous operations from executing without your knowledge.
Leveraging dry-runs
A dry-run means previewing outcomes without actually executing them. It is like trying on clothes before purchasing: if you like them, you buy; if not, you don't.
When asking an AI agent to clean files, rename folders, or delete items, saying Show me in dry-run mode first causes the AI to present a preview: I will move this file here and delete that file there. If the plan looks good, say Proceed; if something looks off, say Cancel.
I want to clean up files older than 90 days in the Downloads folder. Show me a dry-run list of target files first. I will review it before approving execution.
→ The AI lists files targeted for deletion first. Seeing and confirming can these files truly be removed? with your own eyes prevents accidental deletion of crucial data.
24.4 A Unified Real-World Routine Combining All Three Habits
So far, we explored recording, verifying, and preparing rollbacks. These three habits work better when combined into one routine rather than used separately.
Below is an integrated prompt containing all three habits. Make it a routine to use this structure when directing AI agents.
Please perform [Task Description].
1) Record the current state before starting.
2) Create a backup first so changes can be reverted.
3) Change only one thing at a time, comparing before and after numerically at each step.
4) When finished, record the date, changes made, before-and-after comparison, and rollback procedures into the work log.
→ This single prompt includes recording (steps 1 and 4), verification (step 3), and rollback readiness (step 2). Change only the [Task Description] part, and you can use it as a general template for almost any task.
Typing this long prompt each time might feel cumbersome at first. Simply save it in a text file to copy and paste. Alternatively, using the Automation pattern learned in Chapter 22 lets you automate this routine itself.
24.5 Monthly Checkup: Examining Your Computer Once a Month
Just as people undergo an annual health checkup, computers benefit from regular checkups. Once a month is plenty.
There are five checkup items:
First: Is there sufficient free disk space? Having more than 20% free space ensures smooth operation.
Second: Are backups running reliably? Verify the last backup date and its success status.
Third: Are pending updates piling up? Check for OS and application updates.
Fourth: Is SSD health normal? Review S.M.A.R.T. metrics.
Fifth: Are boot times consistent with past benchmarks? If noticeably slower, audit startup programs again.
Checking all five manually takes time, but delegating it to an AI takes only a few minutes.
Run a monthly checkup on my computer.
Check free disk space, last backup date, pending updates, SSD health, and boot times, compiling findings into a health report.
Suggest remedies for any problematic items.
→ The AI audits all five areas at once and produces a report like Free disk space: 23% — Good, SSD Health: 97% — Good. If issues appear, it proposes remediation steps.
Create a scheduled task that automatically runs a system checkup on the 1st of every month and saves results to the AI_Work_Logs folder.
→ The AI writes a monthly audit script and registers it to run on the 1st of every month. Accumulating monthly reports lets you track computer health trends over time at a glance.
[Figure: Example monthly checkup report screen showing Disk, Backup, Updates, SSD, and Boot Time status in a Green/Yellow/Red table]
24.6 The Prompt Savings Account: Collecting High-Performing Prompts
Just as you save money in a bank, gathering effective prompts creates a valuable asset. Let us call this your Prompt Savings Account.
Whenever an AI request produces great results, save that prompt in a dedicated file. When similar situations occur, pull out the saved prompt and use it directly or tweak it slightly.
The structure of a Prompt Savings Account is simple: organize entries in a single text file like this:
Name: Automated Downloads Folder Cleanup
Context: Hundreds of files piled up in Downloads folder
Prompt: "Create a script that automatically sorts files in Downloads into Documents, Images, Archives, and Installers. Show a preview before actually moving files."
Outcome: Automatically categorized into 4 folders. Took 5 minutes.
Appendix 2 of this book includes a ready-to-use prompt collection. Use those prompts as a starting point, adapting them to your needs as you build your own Prompt Savings Account.
Summarize the successful prompts used in this conversation so far. Create a table with Purpose, Original Prompt, and Summary of Results for each.
→ The AI reviews the conversation history and compiles successful prompts into a structured table ready to paste into your Prompt Savings Account.
24.7 Learning from Mistakes
Mistakes are not bad things. A mistake with a record becomes a great teacher that prevents you from repeating it. Only unrecorded mistakes are truly harmful.
Working with AI agents can lead to occasional hiccups: vague prompts generating unexpected results, modifying settings without backups escalating issues, or executing suggested commands without review leading to deleted files.
When this happens, record a Failure Log in your work log. Noting this prompt caused this mistake helps you sidestep the exact same trap next time.
The task just executed produced unexpected results. Analyze the cause and suggest how to refine the prompt to prevent the same mistake next time.
→ The AI analyzes root failure causes and proposes improved prompt phrasing—such as "Because the prompt omitted the target folder path, the wrong folder was modified. Be sure to include the explicit path next time."
24.8 Accumulated Habits Become Mastery
Having read this far, you will have realized that this chapter is not about difficult technical tricks.
The habit of recording is simply appending Log this in the journal when done to your prompt.
The habit of verifying is simply appending Compare before and after in numbers to your prompt.
The habit of preparing rollbacks is simply appending Create a backup before modifying to your prompt.
Habitually embedding these three sentences into your prompts will visibly elevate your computer management skills.
At first, you must add them consciously. But after ten or twenty iterations, a prompt will feel incomplete without them. That is the exact moment when habit transforms into mastery.
Let us recap the five promises from Chapter 4 once more:
1. Measurement comes before feeling.
2. Change only one setting at a time.
3. Preserve pre-change states and rollback methods.
4. Back up before undertaking risky operations like delete or format.
5. Confirm resolution through numbers.
These five promises have served as guiding principles throughout this entire book: from first meeting AI agents in Chapter 1, organizing files in Chapter 6, finding the culprits behind slow PCs in Chapter 14, installing Linux on an Intel Mac in Chapter 17, to mastering prompt patterns in Chapter 22.
If you now practice these five promises naturally without conscious effort, you are already an AI agent power user.
Troubleshooting when things don't work
When log files grow too large: Split files monthly, such as `log-2026-08.md` and `log-2026-09.md` for easy management.
When AI does not know where to save the work log: Specify the explicit folder path in the prompt, such as Save to C:\Users\Name\Documents\AI_Work_Logs\.
When you forget where rollback files are stored: Make it a habit to log rollback file paths directly in the work log entry: Rollback file: C:\Backup\dns-backup-0815.txt.
When monthly scheduled tasks fail to run: Tasks will not execute if the computer is turned off at the scheduled time. Add the condition: If the computer was powered off at the scheduled time, execute upon next boot.
When prompt collections become cluttered: Define categorization criteria, dividing prompts into folders like File Organization, Performance Optimization, and Troubleshooting.
Key takeaway
The secret to using AI agents well is not a special technique, but small daily habits. Record actions, verify with numbers, and prepare rollbacks. Once these three become second nature, you are already a power user.