Human Actually

What happens when algorithms know us better than we know ourselves, and who gets to decide what to do about it.

About

I'm Yuliya. I'm interested in AI governance, digital privacy, online safety, the philosophy of knowledge, and what happens to human agency when the systems we build start making decisions for us. I've spent more than a decade working on trust as an engineering problem, from enterprise security to AI data governance, and every experiment on this site is something I designed and shipped myself.

Human Actually is where I think out loud about all of it. The gap between what technology promises and what it delivers. The line between empowerment and extraction. The epistemological mess of living in a world where seeing is no longer believing and truth has become a team sport.

A lot of this borders on philosophy, and I don't shy away from that. The questions worth asking about AI, privacy, and safety are rarely just technical. They're about what we value, what we're willing to trade, and who we trust to hold the scales.

These are my own views, my own questions, my own tools.

Featured Research Projects

Experiments

Interactive research tools built to test, question, and sharpen judgment in an AI-shaped world.

A growing lab of tools and experiments about trust, verification, and human judgment in the age of AI.

Policy Tracker

AI Policy Pulse

Tracking how AI policy is evolving across jurisdictions.

AI policy is emerging through legislation, regulation, court decisions, standards, and government guidance around the world. Policy Pulse brings those developments together into a single research resource, organizing them by jurisdiction and verifying significant changes through a monthly research cycle.

  • Jurisdiction-level policy tracking
  • Primary-source references and context
  • Monthly research and verification
Explore Policy Pulse
Completed Experiment

The Calibration Cup

A research experiment about confidence, uncertainty, and the quality of human judgment.

During the 2026 FIFA World Cup, I recorded predictions before each match to study whether confidence aligned with reality. The completed experiment evaluates calibration using Brier Score, Calibration Score, and Trust Receipts, with the full methodology and final results published.

  • 104 tournament matches analyzed
  • Final results published
  • Complete prediction history, methodology, and analysis
View Final Results
Research Beta

Trust Dossier

A forensic fact-checking workspace. Not a truth oracle.

Submit a claim, a URL, or a screenshot. Seven specialized AI agents hunt for corroboration, check provenance, build timelines, and run adversarial red-team analysis. You receive a dossier of evidence with explicit uncertainty labels—not a confidence score pretending to be certainty.

  • Uncertainty made legible: inconclusive when the evidence is thin, never performative precision.
  • Your work stays in the browser—no server-side investigation history.
  • SHA-256 integrity seal on every dossier; scope and limits stated plainly.
Open Trust Dossier
Interactive Drill

Deepfake Family Drill

A privacy-first rehearsal for families facing AI-enabled deception.

Deepfakes are not only a media problem. They are becoming a trust problem inside ordinary family life: fake voices, urgent impersonation, hijacked context, pressure before verification. This drill helps people rehearse realistic scenarios, identify weak points, and leave with a practical response protocol.

  • Scenario-based rehearsal, not abstract advice
  • Personalized family protocol with printable version
  • Field Guide with shareable assets and verification workflows
Run the drill
Trust Calibration Experiment

Can You Trust AI?

A judgment test about when AI is safe to trust, when it should be verified, and when human judgment should stay in charge.

AI problems are not only model problems. They are also human trust problems. This experiment helps people see their own pattern.

  • Trust calibration profile
  • What tripped you up
  • A practical framework for using AI better
Open the experiment
Newsletter

The writing

Long-form essays on AI governance, digital privacy, online safety, and the human cost of algorithmic decision-making.

Latest essays

  1. I Built a World Cup Experiment to Keep Receipts on My Predictions (And you should try it too!)
    11 JUNE 2026

    Picking winners is the easy part. The Calibration Cup is a public test of whether the confidence behind each pick was worth anything.

  2. I built an interactive map of global AI policy
    12 MAY 2026

    An experiment to figure out which AI laws actually matter before they matter, with help from two AI agents and a healthy distrust of their output.

  3. The Deepfake Fire Drill Every Family Needs (And Why Your Family Posts Are Now Training Data)
    24 APRIL 2026

    What every household needs to know about voice cloning, deepfake scams, and the plan that works.

  4. I Built a 7-Agent AI Pipeline to Fact-Check a Moon Landing Conspiracy. Here’s What Happened Next.
    9 APRIL 2026

    On "What is truth, really?"

All essays →
What I Cover

Territory

AI Governance

Who audits the auditors? Policy, accountability, and the rules we haven't written yet.

Digital Privacy

Data sovereignty, consent theater, and what "nothing to hide" actually costs you.

Online Safety

Misinformation, deepfakes, platform incentives, and the verification gap.

Human Agency

Algorithmic nudges, attention economics, and the fight to keep thinking for yourself.

Epistemology of AI

What counts as knowledge when machines generate it? Truth, trust, and the limits of certainty.

Ethics & Philosophy

The moral weight of automation. What we owe each other in a world that optimizes for engagement.