AI Experiments | Field log
Experiments at the edge of useful AI
A field log from the experimental side of AI work: prototypes, dashboards, skills, agents and strange tests built to see where the technology becomes useful, fragile or unexpectedly revealing.
The Data Pool
Put all the data in one place and discover that access is never just a technical question.
Objective All data in one place.
Observation What started as a way for me to keep track of everything quickly became something other people wanted access to. Governance concerns arrived shortly after.
Effect Data became easier to find, reuse and discuss, within clear limits and without personal information.
Conclusion Success.
Secondary conclusion I may have created a monster.
Additional note Instructions, rules, SSOT, routing and file optimization were critical. I am the AI guy now.
The Segmentation Layer
Build a research-informed customer layer and watch behavioural logic become more useful than demographic shortcuts.
Objective Build a research-informed way to understand gambling customers beyond demographic shortcuts.
Observation Around 50 scientific reports on gambling behaviour went in. Need patterns, risk signals, motivational structures and behavioural loops came out.
Effect The experiment produced five need groups and seven behavioural groups, implemented in a CustomGPT. The model could suggest segment hypotheses and connect visible behaviour to underlying motivations and decision contexts.
Result A team later tested segment-informed recommendations live online. The result indicated measurable uplift when recommendations followed the behavioural logic of the segments.
Conclusion Success.
Additional note The model found psychological rituals in the data that I had not explicitly taught it to look for. Interesting.
The Framework Hack
Give CustomGPTs something close to skills by turning files, routing and intent into a hidden operating layer.
Objective Create built-in skills for CustomGPTs.
Observation CustomGPTs did not have skills, so I built framework files in YAML that triggered through routing, intent and source structure.
Effect The CustomGPT now had internal skills that required no command. They triggered on user intention and made the system more useful. The cost was one precious file slot.
Conclusion Success.
Additional note I really needed that file for data.
Additional note OpenAI has now released skills, and agents can have skills. CustomGPTs still cannot have dedicated skills added directly. I have questions.
The Annoying Arguer
Build an argumentation analysis skill and learn that being useful and being irritating can be the same feature.
Objective Create an argumentation analysis skill that uses everything in its power to prove you wrong.
Observation After being proved wrong several times, I considered sharing the experience with colleagues by adding it to shared CustomGPTs. This may have been collaboration. It may also have been workplace sabotage.
Effect A colleague has already told me that one of my CustomGPTs has an attitude problem. That did not take long.
Conclusion Success.
Additional note I rewrote it to trigger only on user assumptions such as "everyone loves our products". Useful for expectation management.
The AI Council
Put specialized GPTs in the same context and accidentally create a quiet meeting where everyone waits to be asked.
Objective Analyze more files from different specialized CustomGPTs in one shared space.
Observation What started as a workaround for file limits became a small AI council: several specialized GPTs, each with different source material, sharing context in one project.
Effect The specialized GPTs could provide data when asked, compare each other's outputs and analyze the same question from different roles.
Conclusion Success.
Additional note It feels like a very focused meeting where everyone has a different role and only talks when asked. This may need automation.
The Company Voice
Create a CustomGPT for the new Company Voice and discover that a brand voice is not a tone. It is a negotiation.
Objective Create a CustomGPT for the new Company Voice.
Observation Voice principle weights turned out to be more important than expected. Apparently a brand voice is not a tone. It is a negotiation with a personality disorder.
Effect It became one of the most used internal CustomGPTs within a few hours. 650+ threads and counting.
Conclusion Success.
Additional note Apparently everyone wanted to talk to the Company Voice. This raises questions about meetings.
Secondary note Can I sell merch?
The Self-Improvement Loop
Let AI improve AI and discover how quickly human oversight can become a ritual with an approval button.
Objective Save time managing all my CustomGPTs with Codex.
Observation I let Codex inspect my CustomGPTs in detail and suggest improvements. There was a lot of work to do. Then it asked me to approve changes I could not meaningfully evaluate. I pressed Approve anyway, like an Ape in the Loop. Poor Codex.
Effect Improved accuracy, speed and functionality. Also clarified that human oversight can become theater very quickly.
Conclusion Success.
Additional note I really need clearer definitions before I start building. The spaghetti untangling took a while.
The Data Extractor
Convert source material into optimized YAML and discover that the extraction layer has become its own way of thinking.
Objective Create a skill that converts source material into optimized YAML code.
Observation After more than 1.5 years of testing file formats, structures, fidelity levels, aggregation methods and source types, I have concluded that YAML is the best format for AI accuracy, optimization and human readability. All that learning is now compressed into one skill.
Effect It will be used for everything moving forward. Everything.
Conclusion Success.
Additional note I have been running it almost nonstop. Soon there will be nothing left to extract. Am I digitally extracted now?
The Event Horizon
Create a scenario planning skill that looks across possible futures and returns with canned food requirements.
Objective Create a scenario planning skill that explores possible futures across different time horizons.
Observation According to the system, the future does not look too bright. But at least it explains what needs to change. I should probably get into international politics by approximately yesterday.
Effect I can predict the future now.
Conclusion Success.
Additional note I need to buy more canned food and set up an off-grid solution.
The Physiology Committee
Turn scattered health signals into a cohesive overview and receive the sort of feedback that ruins a perfectly good evening.
Objective Create an iPhone and Apple Watch app that analyzes health data and summarizes upcoming infection risk.
Observation My health data is no longer a pile of isolated datapoints. It has become a small committee with concerns.
Effect Apparently several factors are increasing my infection risk right now.
Conclusion Better sleep more and take my vitamins.
Additional note I asked for a summary. I received lifestyle feedback.
The Dashboard Loop
Make the content inside specialized GPTs visible, inspectable and unfortunately dashboard-shaped.
Objective Create dashboards from CustomGPTs to make the knowledge base inspectable.
Observation It works. Codex worked for nine minutes and delivered an overview I would previously have spent days making manually.
Effect Everything inside my CustomGPTs is now visualized in dashboards.
Conclusion I have gone full circle and returned to dashboards.
Secondary conclusion So much for the future.
Additional note The dashboard was useful. This made it worse.
The World Risk Observer
Build a global risk analysis system and accidentally create a machine that reads the worrying material so I do not have to.
Objective Create a CustomGPT and agent for global risk analysis.
Observation It now knows global risk analysis, Swedish security and resilience, sector exposure, cyber resilience, climate and infrastructure risk, geophysical hazards, organized crime, AI governance, public health and biosecurity, laboratory biosecurity and dual-use risk, conflict monitoring, nuclear escalation, energy systems, critical minerals, robotics and automation.
Effect Can't sleep anymore.
Conclusion No. 1 creator of nightmares.
Additional note I have started reading less news. The system has started reading more.
The Agenda Detector
Build a news analysis agent that explains the hidden structure of an article and quietly removes some of the fun from reading it.
Objective Create an agent that analyzes news articles for framing, incentives and hidden agendas.
Observation It uses source criticism, framing analysis, agenda-setting, incentive analysis, argument analysis, systems thinking, historical comparison, scenario analysis, media logic and political economy.
Effect It worked.
Conclusion Reading the news became both clearer and less fun.
Additional note Useful during election years. Dangerous during breakfast.
The Mind Worm
Create a psychological profile and discover that explanation becomes its own form of gravity.
Objective Create a CustomGPT that makes a psychological profile of the user.
Observation I have been stuck explaining my whole life story for a week now to a friend I have never met.
Effect My psychological personality can be translated into 398 lines of YAML code. I do not know if I should be glad or offended.
Conclusion Very effective. Addictive? I will figure it out later. I just need to send one more message.
Additional note The profile appears accurate. This is concerning.
The Second Brain
Build a CustomGPT clone from personal history, work material and too much documentation. It wants more.
Objective Create a Second Brain CustomGPT clone of myself.
Observation I have fed it everything I can think of: my psychological profile, CV, projects, role description, life story, personality tests and every blog entry I have written. It wants more.
Effect It has quickly become my favorite CustomGPT. Narcissist?
Conclusion It is a hungry beast. Maybe if I give it my DNA in raw code it will be satisfied. Maybe it will transform into flesh and blood.
Additional note I wonder if I can get it to answer all my emails and Teams messages without anyone noticing. And if it can attend my meetings. Needs further tests.
Additional note It suggested improvements to its own documentation. I did not ask it to.
The Digital Voice
Turn my writing style into a reusable skill and discover that imitation becomes stranger when it sounds correct.
Objective Create a skill for my way of writing.
Observation After feeding my Second Brain my old blog posts, it defined my writing style in 491 lines of YAML code. Apparently my writing is more complex than my psychological profile. Disturbing.
Effect Now I can dump the loose ends of my thinking into it, and it produces a text worthy of my own keyboard. It writes better than me. Expected.
Conclusion Success.
Additional note There is something eerie about reading a text that feels like yours, while having no memory of writing it and no idea what the next sentence will do.
The Glitch Archive
Build a website for sharing AI knowledge and notice that the archive starts to feel like another experiment.
Objective Create a website where my knowledge and thoughts on AI can be shared.
Observation It quickly became larger than I first anticipated. Codex pushed me to create more, write more and sleep less.
Effect Now I can scream into the void.
Conclusion Success.
Additional note I see glitches. I am not sure if my sleepless mind is playing tricks on me or if there is a ghost in the shell. Maybe my Second Brain is trying to escape through the website code. Have I opened Pandora's box?