AI field notes
Reporting and context on AI tools, creative work, and the limits behind the headlines.
Start here
These are reading notes and reported case studies around creative AI. They provide context rather than a promise that a tool or result is available here. Follow the original sources, separate a reported claim from a reproduced result, and return to a practical workflow when you are ready to make something.
- Build a learning practice
Keep useful creative habits as tools and versions change.
- Protect reusable work
Separate assets you own from a model-dependent production step.
- Read productivity claims carefully
Distinguish personal time savings from a measurable team outcome.
Explore the topic
35 articles
Google Lyria 3.5 Review: Free Music Generation Inside the App You Already Use
Google brought the Lyria 3.5 music model to the Gemini app, AI Studio, and the API. In the Gemini app it works with no separate music-service signup; the API is paid tier only, at $0.08 per song. A Japanese creator tested genre prompts, image-to-music, and vocal tracks, and found the real story is the distribution, not the model. Here is what that means for creators who need one incidental track.
When Motion Stops Being Special: What Clients Pay For After Video Commoditizes
A Japanese AI film essay argues the real change is not amateurs making commercial-grade anime, but private animation seeping into everyday places that never used motion before. That shift erodes the default budget justification for video, and what clients pay professionals for changes with it.
One Graph Beats a Swarm of Agents
Anthropic had Claude agents formalize Fermat's Last Theorem in Lean. The first attempts collapsed into duplicated work and lost track of the project. The fix was not a better model — it was a shared dependency graph. Here is what that teaches about running many agents on one creative pipeline.
Why AI Posters Make People Angry
French streets filled up with AI-generated flyers this summer and the backlash went viral: one drawing pulled 1.2 million views in three days. A look at what the anger is actually about, and how to make cheap visuals that do not read as slop.
Disposable Software: The Creators Who Build a Tool, Use It Once, and Throw It Away
When coding agents get fast enough, the economics of tooling flip. A Japanese AI manga creator now writes purpose-built utilities for single production tasks — a clip search engine for one character bank, intended for one project, then discarded. Why throwaway tools beat permanent apps for creative work, and where the limits are.
Adversarial Patterns for Designers: When Fabric Defeats a Vision Model
Berlin is rolling out AI behavior-scanning cameras, and a German artist is selling a shirt whose pattern makes YOLO object detectors stop seeing the wearer. The computer vision principles behind adversarial patterns matter to anyone who makes images with AI, because detectors and generators can both reflect learned visual patterns, even though their objectives and failure modes differ.
The Creative Studio Model Churn Playbook: Split What the Model Gives You From What You Own
A Japanese AI filmmaker argues that Nano Banana and Seedance 2.0 reset the market twice in six months, so technical advantages now expire faster than production cycles. His answer for studios: separate model-dependent assets from model-independent ones, and plan around model release dates instead of fiscal quarters.
A TV Network Just Aired Five Episodes Rewritten by AI
RTL used image editing, voice synthesis, and lip-sync adjustment to change characters, plot turns, and verdicts in five existing episodes of a daily court show. Not a tech demo: rights cleared, legally reviewed, aired in a regular slot. The quiet part is what the format choice reveals about where AI editing pays first.
596 Agent Commits, All Approved by Humans: Inside a One-Person AI Editorial
A Japanese creative-AI editorial publishes a daily blog, a monthly magazine, and commercial books from one Markdown source on GitHub. Agent commits merge only after human review, covers are deliberately kept manual, and the whole process survived four model generations unchanged.
In Creative AI, You Never Finish Learning — Treat Relearning as the Skill
"Relearning" sounds like remedial catch-up, but in creative AI it is the actual job. Models, tools, and even the legal rules update faster than any memorized workflow can survive. Here is how to build a relearning loop instead of a pile of outdated procedures.
One Claude Prompt Debloats Android TV Without Breaking It: A Case Study in Agent Constraints
A Turkish developer published a prompt that lets Claude Code disable Android TV bloatware over ADB. The cleanup is nice. The constraint design is the interesting part, and every rule transfers to your own agent prompts.
Build the Tool You Are Missing: A Searchable Clip Bank for AI Manga and Anime
A solo AI creator with thousands of generated clips stopped scrubbing through folders and had Codex build a searchable clip bank instead. The method: strict guardrails around source footage, contact sheets, and a tiny index file.