Trustless Computing Certification Body

(Updated October 9th, 2024)

The Trustless Computing Certification Body Initiative (“Initiative”) is a proposal for a new intergovernmental organization for the creation, regulation and joint exploitation of secure digital communications and control systems for frontier AIs and other critical societal systems. 

The Initiative is aggregating a critical mass of globally-diverse states, intergovernmental organizations and neutral INGOs to build new open, neutral and participatory intergovernmental organizations to develop and certify radically more trustworthy, multi-laterally-trusted and accountable end-to-end IT systems, for use in (1) confidential and diplomatic communications (or Seevik Net) and (2) control subsystems for critical AIs, social media and other critical systems (or Seevik Controls) - while enabling legitimate lawful access, national and international.

While standalone, the Initiative is also one of the three agencies of a new intergovernmental organization proposed by our Harnessing AI Risk Initiative to manage globally AI and digital communications for the global public good, under the provisional name of IT Security Agency - now advanced primarily via the Coalition for a Baruch Plan for AI. 

The Problem with Control Systems for Global Highly-critical Infrastructure and Advanced AIs 

Critical control subsystems are used globally in various civilian and military applications for critical infrastructure, including highly critical cases like nuclear facilities and electrical grids. Some of this infrastructure is transnational or global, with potentially grave international consequences if compromised.

Global highly-critical infrastructure includes international law-abiding confidential communications and social media feed systems, which can significantly impact global peace and democracy if compromised.

Frontier AI models represent a new category of global very-highly-critical infrastructure, posing grave risks of malfunction, attacks, and potential loss of control, with immense implications for human safety and power concentration.

While control systems for Frontier AI models are being implemented using state-of-the-art and military-grade technologies, their actual assurance levels are insufficient to inspire equal confidence across different geopolitical alliances.

The Problem with Global Highly-Sensitive Digital Communications

Heads of states, ministers, diplomats, elected officials, journalists and top scientists do not have access to interoperable computing devices and services that enable them to meaningfully protect the confidentiality of their sensitive and off-the-record communications against abuse by innumerable state and non-state hacking entities. They are therefore forced to rely only or mostly on in-person meetings to further global cooperation and diplomatic initiatives.

The lack of sufficiently secure and interoperable communication-enables computing devices and services is due to a number of factors. One factor is the need to protect national security. Governments are concerned that if they allow their officials to use interoperable computing devices and services, it could open the door to hacking and other forms of cyber attacks. Another factor is the lack of standardization. There is no single standard for interoperable computing devices and services, which makes it difficult for governments to develop and deploy them.

The consequences are significant. It makes it difficult for heads of states and other government officials to communicate with each other in a secure and confidential manner. This can lead to misunderstandings and conflict, as well as making it more difficult to resolve international disputes. It also makes it difficult for journalists to investigate important stories and for scientists to share their research findings.

Our Approach

Key to achieving such actual and perceived trustworthiness is the application of the Trustless Computing Paradigms, which emphasize: 

  1. much higher transparency of technical designs and processes; 

  2. much more expert and varied security reviews in relation to complexity, 

  3. trustworthy procedural in-person legitimate lawful access mechanisms, and most importantly

  4. higher global participation and neutrality in the standardization and certification governance processes.

The Solution

The Initiative entails the creation, via participatory and neutral processes, of:

  • a new intergovernmental IT security certification and governance body, TCCB, and 

  • a set of initial infrastructure and ecosystems complaint with it based on battle-tested and open systems:

    • a new cloud infrastructure and supply chain ecosystem, multinational, neutral and redundant

    • a new modular endpoint platform, for use as client, server and control system for use in two domains:

  • Seevik Net,  a 2mm-thin ultra-secure mobile device, carried in custom leather wallets or embedded in the back of Android, iOS or Harmony smartphones (proof-of-concept device video). Via such form factor, it will initially be running alongside current hegemonic mobile devices, apps and cloud services, instead of trying to replace them. 

  • Seevik Controls, a base control system endpoint platform, suitable for critical functions related to compliance systems, access controls, security monitoring, firmware upgrade and AI value systems.

Our "pivot" of March 2023

Until March 2023, Seevik Net was the first "beach-head" use case to generate the demand that would then lead to develop Seevik Controls.

The hesitations of states to face with their lack of access to secure international critical communications (due to the "ostrich effect" or "glass house effect"), and the explosion of AI capabilities and related demand for radically more actual and perceived trustworthiness of AI system, have lead us to put forward Seevik Controls for critical AI system as the first "beach-head" use case, and expand our vision for intergovernmental organization to other functions as outlined in the Harnessing AI Risk Initiative, now primarily advanced via the Coalition for a Baruch Plan for AI.

The plan that follows has yet to be fully updated since such a pivot, so you'll find more emphasis on Seevik Net rather than Seevik Controls.

1.8. Progress & Traction

As detailed 25-pages Traction Update, in addition to nine states and three IGOs interested to join the Initiative, high officials of six states from Africa, Asia and South America, including three Permanent Representatives of their Missions to the UN in Geneva, have participated as prospective partners in the 9th and 10th editions of the Free and Safe in Cyberspace event series we held in Geneva in Q2 2023.

So far, we've built the high-level technical, socio-technical and governance details of the certification body, and launched it in 2021 in the presence of prestigious personalities. We built a proof-of-concept device of a TCCB-compliant mobile client-side endpoint device (video). We have designed a large part of the supply chain and ecosystem, while the rest will be completed with strategic firms from participating states.

Roadmap

We’ll be holding several preparatory meetings leading up to the 1st Harnessing AI Risk Summit and possibly future editions of the Free and Safe in Cyberspace, in Geneva.

We aim soon after then to select a limited number  globally-diverse states, IGOs and NGOs to become the initial cofounder partners (at an average cost of CHF 2,000,000 each) of such open intergovernmental joint-venture to build and govern the 1st global democratic digital communications infrastructure, to be extended to Seevik Controls. A few more will be selected to join as mere governance partners with smaller yearly fees.

Neutrality

Given how central neutrality is to our Initiative, the current cyber superpowers of the USA, China and Israel are very highly welcome to join the Initiative as cofounder or governance partners and join FSC12 but their application will be held in suspension until all three of them will have applied (as you can read in chapter 13 of our Executive Summary, or here on our site).

Why and How Lawful Access?

Since the commercial availability of those levels of security have consistently and structurally proven to be incompatible with global public safety, the same extreme safeguards that ensure radically higher trustworthiness will be applied to applied to ensure sufficiently secure mechanisms for their in-person procedural legitimate lawful access, national and international, within the confines of current national and international laws.

More Information

Refer to the:

  • 30-page Master Slides (Google Slides)

  • 80-page Trustless Computing Certification Body Master Plan (PDF)