AIR: Platform Mitigating X-Risk through Human Capacity Ops

I am realizing my research on X-Risk by engineering an infrastructure aiming to connect all AI Safety labs across the world with talent from all walks of life and backgrounds, allowing them to streamline HR and R&D Management Operations and amplify their impact through collaboration, not competition.

The highest risk from Transformative AI (TAI) is power concentration, and its effects escalate today, not only through our inability to address technical loss-of-control failures, but most importantly through defective social governance structures. Whether these are inefficient and biased government protocols, or discriminative business processes, human capacity must be empowered proactively to provide continuous oversight of AI.

I founded my lab during the Greek economic crisis, operating from a context of systemic constraint and real existential risk from misuse of power, shaping a research program for the past 8 years aiming to build and democratize safety infrastructure. This fund will allow me to deliver an open-source local-first app, connecting AI Safety Labs through a decentralized network, allowing them to scale talent and operations through shared, verifiable coordination standards that I have already developed.

While AIR is the deliverable infrastructure, it is based on a robust epistemic taxonomy of AI Safety risks which addresses explicitly x-Risk. It will also include a built-in Plan-B economy that will initially be used as a verification system for alignment work, and over time has the capacity to replace our collapsing debt-based economic system with a physically abundant and verifiable capacity (the atomic frequency - the same standard that we currently use for coordination of our clocks and GPS systems).
https://gyrogovernance.com/#thm
https://gyrogovernance.com/#air
https://gyrogovernance.com/#moments

The core thesis is that x-risk from AI is not merely a technical capability problem, but a structural governance and economic failure. Specifically, it is the systemic displacement of human governance (treating Indirect AI systems as Direct sources of authority and agency) sustained long enough that human capacity to recover control atrophies past a civilizational point of no return. THM, AIR, and the Moments Economy form a complete stack to prevent this:THM provides the diagnosis and rules, AIR provides the verifiable logistical infrastructure, and the Moments Economy flips the economic incentives from displacing humans to valuing human oversight.

The theoretical foundations, prototype, and evaluation frameworks are already developed, tested, and documented. A 12 month period allows for immediate product development, a pilot with partner labs, and early adoption.

A $50,000 grant would allow me to work on the project solo. A $500,000 grant would allow me to hire a small team so I can deliver faster and communicate and support adoption in a more sustainable way.

The ideal structure is for a lab or aligned organization to host and hire me, and provide the basic operational resources, so I can lead a small team and build AIR inside an existing AI safety environment. This reduces overhead, lowers delivery risk, and improves ROI, since many labs already budget for most of the underlying costs and have the right context for pilot deployment. If you can fund my placement in a lab within your network, in a Principal AI Safety Design Engineer role, I can lead the build and rollout with the best resource efficiency for everyone.
\

AIR Budget Plan (12 months)\

\

Funding levels and what changes\

**Minimum: $50,000**\

• Goal: Build and ship a usable pilot version on a solo basis\
• Success criteria:\
  • 1 working local-first app prototype released open source\
  • 1 to 2 labs complete a pilot run using it\
  • At least 20 verified work submissions routed through the workflow
    
    **Standard: $150,000**\
• Goal: Build faster and run a real pilot with more labs\
• Success criteria:\
  • Stable beta release plus onboarding docs\
  • 3 to 5 labs complete pilot runs\
  • At least 100 verified work submissions\
  • At least 2 repeat projects run by partner labs
    
    **Optimal: $500,000**\
• Goal: Deliver a production-ready app plus adoption support and training\
• Success criteria:\
  • Production-ready v1 release with upgrades based on pilot feedback\
  • 10 or more labs onboarded\
  • At least 500 verified work submissions\
  • At least 10 completed lab projects run end to end through the system
    
    More funding increases delivery speed and adoption capacity. It does not change the core mission.
    
    ---
    \

Optimal: Budget Breakdown: $500,000\

\

1) Staff: $390,000\

• Research Director (me, full time): $140,000
  Product direction, research-to-product translation, lab partnerships\
• Operations Lead (full time): $95,000
  Lab onboarding, program execution, grant admin, fiscal host coordination\
• Software Engineer (full time): $90,000
  App build, local-first sync, reliability, releases, maintenance\
• Community and Training Coordinator (full time): $65,000
  Contributor onboarding, training delivery, pilot support, communications
  \

2) Programs: $45,000\

• Compute and API costs: $15,000
  Testing integrations and running evaluation-style pilots with labs\
• Pilot support and participant stipends: $20,000
  Targeted support to ensure pilot runs complete with usable feedback\
• Workshops and outreach: $10,000
  Onboarding sessions for labs and new contributors, mostly online
  \

3) Admin and compliance: $40,000\

• Fiscal sponsor and compliance costs: $35,000
  Assumes a 7 percent fiscal sponsorship fee on $500,000\
• Legal and accounting: $3,000
  Contracts, compliance review\
• Tools and basic operations: $2,000
  Hosting, domains, communications, repo tooling
  \

4) Buffer: $25,000\

• Contingency for currency fluctuations, unexpected tooling needs, or pilot expansion
  
  ---
  \

Total: $500,000\

• Staff: $390,000\
• Programs: $45,000\
• Admin and compliance: $40,000\
• Buffer: $25,000
  
  ---
  \

Short justification\

The research and core framework are already developed. This budget funds execution: building the app, running pilots with real labs, onboarding contributors, and supporting adoption. The main cost is people because year 1 is an engineering and rollout year.