The future requires human signal to overcome artificial noise. Frameworks for institutional operators governing AI-disrupted systems.
The Platform
Founded by Dr. Tuboise Floyd
Editor in Chief, The AI Governance Record · Host, The AI Governance Briefing with Dr. Tuboise Floyd · TAIMScore™ Certified Assessor
Canonical IP · Human Signal
Each framework is a standalone diagnostic or operational tool for institutional operators governing AI.
Analysis
Two levels of institutional AI governance failure. Structural absence. Structural insufficiency. Permitted is not the same as admissible.
Read the framework ›Diagnostic
Governance As a Structural Problem. Most institutions don't lack the right software. They never built the right structure.
Read the diagnostic ›Thesis
Institutions deploying AI fail not because of underperforming models, but because of broken governance structures. The primary risk is never a bad model.
Read the thesis ›Practice
The algorithm is rewriting your source code. Cognitive defense for operators drowning in vendor hype and feed-induced source amnesia.
Read the brief ›Framework
Four physical constraints every AI strategy must address — Lithography, Energy, Arbitrage, Cooling. Miss one, you are leaking value.
Architecture
Presence Signaling Architecture and AI as Presence Interface — frameworks for restoring human visibility in systems designed to observe, not listen.
Read the architecture ›Signal Validation
An emergent lexicon entry from the PSA® runtime. The fusion of a scored identity-coded signal and a prompt — a presence-optimized query.
Read the lexicon ›Live Diagnostic · GASP™
5 questions. Instant GASP™ structural diagnosis. No login. Find out if your institution has the structure to govern AI accountably — right now.
Start ›TAIMScore™ in action
Human Signal applies the TAIMScore™ framework to real AI failures on the podcast. 12 incidents. 12 TAIM controls. Every failure is a practice scenario.
FAILURE FILE 01 / 12
Accountability & Training
Microsoft TAY · AIID #6
Microsoft spent $0 on adversarial input controls before releasing TAY. Within 16 hours it published racist propaganda. Four TAIM domains failed simultaneously.
FAILURE FILE 05 / 12
Privacy & Socio-Technical Design
OpenAI Class Action · AIID #561
A 157-page class action alleged ChatGPT was trained on private data without consent. Active exposure under HIPAA, TRAIGA, and the EU AI Act simultaneously.
FAILURE FILE 09 / 12
Bias, Fairness & Contextual Deployment
Wrongful Arrests · AIID #74 · #896
Three Black men. Three wrongful arrests. Detroit Police acknowledged a 96% misidentification rate when used in isolation. Detroit settled for $300,000.
FAILURE FILE 12 / 12
Feedback Systems & Context-Appropriate Use
Vanderbilt / ChatGPT
Vanderbilt sent a post-mass-shooting condolence email signed "Paraphrase from OpenAI's ChatGPT." No mechanism asked: must a human own these words?
April 2026 · Failure File of the Month
Air Canada · BC Civil Resolution Tribunal · 2024
Air Canada's chatbot told a grieving customer he could apply for a bereavement fare retroactively within 90 days. That policy didn't exist. When he submitted the claim, Air Canada denied it — and argued the chatbot was a separate legal entity not binding on the airline. The Tribunal rejected that defense and awarded the customer $812.
The precedent: you own what your AI says.
Read full analysisNow broadcasting
Rapid-fire episodes on AI governance, institutional risk, and finding your human value when the machine noise gets loud.
AI Governance
Finding your human value when AI rewrites the rules. Why "leverage" is replacing judgment.
BAR Method
Background, Action, Result as a personal reinvention framework. The ultimate self-interview.
Identity
Five steps to break autopilot, claim solitude, and find purpose through serving others.
Career
How market signals shape opportunity. Stand out by amplifying your story at the right moment.
50 seats · 8 experts · No vendor scripts · One hour of unfiltered diagnosis — with your questions at the center.
SESSION STRUCTURE · 60 MIN
GENERAL ADMISSION
$147 / seat
Secure checkout via Stripe. Can't attend live? You still get the recording.
Quick reference
TAIMScore™ — the Trusted AI Model Score — is an enterprise AI maturity and risk assessment framework developed by HISPI Project Cerebellum. It gives auditors, executives, and compliance professionals a structured methodology to score, audit, and manage an organization's AI readiness. Human Signal is an authorized affiliate partner promoting the official TAIMScore™ Assessor Workshop — virtual, hands-on, 6 CPEs.
Founder of Human Signal and an independent AI governance researcher. He developed the LEAC Protocol and the Noise Discipline Framework — tools for restoring human visibility and institutional signal in automated, high-noise environments. Hosts The AI Governance Briefing with Dr. Tuboise Floyd and leads the quarterly Town Hall for institutional operators.
A macro diagnostic tool built from forensic market analysis — not a governance framework. It identifies the physical infrastructure constraints that determine which AI companies survive the infrastructure war and where value erodes.
The AI Governance Briefing with Dr. Tuboise Floyd is an AI governance podcast for the Builder Class — leaders, auditors, and institutional operators navigating AI-disrupted systems. Dr. Floyd examines the physics of institutional failure, the limits of automation, and what it takes to govern the machine.
Three underwriting tiers — Signal Drop Sponsorships ($1,500 per episode), Seasonal Signal Partnerships ($6,000 per quarter), and the Signal Brief Presenting Partner package ($12,000 per quarter), which includes named sponsorship of the Quarterly Town Hall and direct introductions to Dr. Floyd's institutional network.
Visible Human Community
Failure Files™ digest, monthly session reminders, and AI governance framework updates. Join practitioners, auditors, and governance leaders who refuse to let the machine win.
Take the pledgeWe do not sell your data.
