We value your privacy.
Airtrek AI uses essential cookies for the app to work and optional analytics cookies (Google Analytics) to understand how people use the site. We do not sell your data. Read more in our Privacy Policy.
Thoughts on AI, culture, neuroscience, and building the world's first virtual time machine
New to Airtrek? Meet Lyra, your AI travel guide. Chat with era-locked characters, hear them speak, transform your selfie, collect passport stamps, create your own characters, and request historical figures. The complete visual guide.
Read more →
The Swahili Coast and Great Zimbabwe were medieval African civilizations with coral-stone mosques, dry-stone walls taller than any in sub-Saharan Africa, and trade networks reaching China. Most people will never hear about them. Here is what I found when I visited both on Airtrek.
In 1880 New York, a railroad magnate, an Irish immigrant servant, and a Tammany Hall ward boss showed me three sides of a technological revolution. The parallels to today are not a metaphor. They are a pattern.
In 1176 CE Chengdu, the world's first paper money was circulating. A scholar-official, a tea merchant, and an exchange clerk gave me three completely different answers about what money actually is. The answer is never just one thing.
In Tikal, 750 CE, I asked three people the same question: what keeps the world from falling apart? A divine lord said ritual. A scribe said memory. A farmer said the milpa. They were all describing the same world from three positions inside it.
Every AI system gives you one answer. Airtrek gives you three contradictions and asks you to think. Every claim is backed by clickable sources. And the eras we build next are the ones you ask for, built with cultural experts who are paid for their knowledge.
When students use AI to write the essay, they skip the thinking that education is supposed to develop. The answer is not to ban AI. It is to distinguish between AI that replaces thinking and AI that creates conditions for better thinking.
I asked Homer, Odysseus, and Penelope, all alive in 8th-century BCE Homeric Greece, the same question about what outlasts a person. Their answers weren't just different. They were three different ways of being alive. This is why one AI historian is not enough.
I uploaded a selfie and arrived in 110 CE as a Roman citizen. Not a filter, a portrait that put me in the scene. Then I talked to the people who lived there. Here's what it felt like to see myself in the past.
The hardest part of building Airtrek isn't the 3D or the voices. It's making characters who only know their own time, hold views that clash with ours, and admit when the record runs out. Here's what we've learned building AI people of the past.
I used Airtrek to travel to Renaissance Florence and asked Michelangelo what he felt the first time a form emerged from the marble. No letter, no diary, no contemporary account records it. Here's what happened when I asked an AI character built to be honest about what the record doesn't preserve.
Egypt under Ramesses II is live at Airtrek.ai to challenge your conceptions of ancient life, and remind us of what the ancients knew that we have forgotten.
Explore the past instead of only reading about it. Step into an era, see yourself in it, and talk with people from another time, not just read about them.
AI historical figures that go beyond a static chatbot. Talk with people from another era inside their world, multiple perspectives, not one narrator.
Step into Imperial Rome, 110 CE, under Emperor Trajan. Talk with a senator, a merchant, a noblewoman, a Vestal Virgin. Explore the past, don't just read it.
AI characters with era-locked knowledge, distinct voices, selfie-to-era portraits, persistent memory, and honest handling of uncertainty.
How Airtrek handles historical accuracy, uncertainty, anachronism, and creative reconstruction. A trust asset, not a disclaimer.
Women bring a desperately needed perspective to AI, rigorous attention to cultural detail that current teams have overlooked. And I have the proof.
We built a 75-question benchmark focused on Rome in 110 CE and compared a raw Mistral Small model to Airtrek's Triad Engine. Hallucinations dropped from 61% to 21%.
UBI addresses the reality that automation displaces labor faster than new jobs appear. The Airtrek DAO model flips both of UBI's weaknesses, it's non-extractive and active.
VR simulated environments don't just trick your eyes, they hit your brain like a freight train, firing up neuroplasticity, motor circuits, and peak performance pathways.
Most AI today is sanitized by design. That's great for brand safety. It's terrible for human growth. Airtrek's approach is the opposite of sanitized AI.
The internet was built by one culture. AI learned from it. And now we're surprised it doesn't understand the rest of humanity?
If crash dummies had male bodies for decades, what do today's AI training datasets get wrong about the cultures, voices, and knowledge of 80% of humanity?
Airtrek replaces one all-knowing AI chatbot with three interwoven character sets, everyday locals, wise guides, reflective mirrors, forcing users to compare views, question biases, and synthesize truths. This trains real critical thinking instead of outsourcing judgment to single-stream agents.
After months of building, the world's first web-based virtual time machine is online. Pilot a 3D cockpit portal, explore immersive eras and characters.
Ever wondered how you'd appear as a Roman citizen? Our AI-powered virtual time machine transforms your selfie into a historically accurate Roman avatar.
Meet Marcus the Senator, Julia the Noble, and other fascinating characters from Ancient Rome. Have real AI-powered conversations with history.
