Runtime authorization infrastructure for AI in regulated industries.
Spanning semantic, constraint, causal, adaptive, and consensus determinism, with cross-cutting assurance mechanisms.
FERZ builds runtime authorization infrastructure: the enforcement layer of AI governance for regulated industries. The architecture is organized around five pathway engines plus cross-cutting assurance mechanisms, with governed actions recorded through tamper-evident authorization artifacts in a common audit-record format. Designed for deployment in healthcare, financial services, and government and defense, where operational consequence is bounded by regulatory and statutory requirements.
AI outputs, decisions, actions
→
Runtime authorization boundary
Fail-closed
→
Five determinism pathways and cross-cutting assurance
→
Tamper-evident authorization artifacts
→
Regulated execution
Interface view. Monitoring can inform a verdict; it cannot substitute for one.
Deterministic assurance addresses a structural requirement that current AI architectures do not satisfy.
Semantic determinism
LASO(f)
Constraint determinism
DELIA
Causal determinism
CausaCore
Adaptive determinism
Behavioral Engine
Consensus determinism
Constitutional Blockchain
Evaluation Framework
How FERZ is evaluated.
FERZ is evaluated differently than application-layer AI products because the primary assets are architectural pathways, intellectual property, standards position, and regulatory alignment.
Investors typically evaluate FERZ across:
Architectural pathway coverage
Five enforcement architectures under filed patent applications, each independently licensable, each addressing a distinct deterministic-assurance mechanism.
Intellectual property depth
Patent filings across the five pathways, plus defensive publications establishing prior art, a published standard, and trade-secret protection on additional implementation detail.
Standards position
The Five Tests Standard (5TS) published as a vendor-neutral standard at github.com/edmeyman/4ts-standard.
Research foundation
More than 50 published research artifacts across Zenodo, SSRN, technical articles, standards, and defensive-publication records, establishing the formal case for deterministic assurance and defining its architectural primitives.
Regulatory timing
Positioned ahead of the documentation and accountability requirements that current regulatory frameworks are beginning to formalize, with patent filings, publications, and standards already in place.
Why Now
Regulated industries cannot deploy AI in core processes without enforceable assurance.
Drug approval pipelines, lending decisions, clinical workflows, autonomous system authorizations: these require assurance that produces the same decision under identical conditions, fails closed when it cannot decide, and generates evidence artifacts that survive regulatory audit. This is a structural requirement, not a preference.
Existing approaches provide visibility, alignment, or post-hoc controls. FERZ focuses on deterministic assurance before execution.
Per-engine guardrails are probabilistic and weaken under multi-agent orchestration. Human-in-the-loop preserves the appearance of AI adoption while reducing its economic case. These are architectural ceilings, not implementation gaps.
Current Approaches
Probabilistic guardrails
Post-hoc monitoring
Human-in-the-loop bottlenecks
Optional middleware layers
Logs as governance artifacts
Observability after execution
What Is Required
Pre-execution authorization gates
Non-bypassable enforcement boundaries
Fail-closed authorization architecture
Tamper-evident authorization artifacts
Runtime constraint enforcement
Ex-ante authorization before execution
Applied Theory
Published science. Patent-pending architecture. Deployable infrastructure.
FERZ is not a research project and it is not a wrapper. The company published the theoretical foundations for deterministic assurance, then filed patent applications on the architectural pathways those foundations require, and is building the runtime infrastructure that implements them. Each layer of the company traces back to the one before it. This is infrastructure for the portion of the AI market that cannot operate without enforced assurance.
01: Theory
A published research corpus spanning DOI-archived papers, standards, technical notes, technical articles, and defensive-publication records. Establishes the formal case for deterministic assurance and defines its architectural primitives.
50+ research artifacts
Published research corpus
02: Protection
Patent filings across multiple independent enforcement architectures. Defensive publications establishing prior art. A published standard positioned for regulatory citation. Trade-secret protection on additional implementation detail. Competing implementations must navigate a dense IP and architecture landscape shaped by FERZ's filings, publications, and standards.
5 families
On the international PCT track
03: Infrastructure
Deterministic infrastructure engineered from the patent-pending architectures. Integrates where AI outputs, decisions, and actions become operationally consequential.
3 verticals
Healthcare, finserv, gov/def
5
PCT-Track Patent Families
4
Defensive Publications
1
Published Standard
IP coverage spans multiple independent enforcement architectures. Competing implementations must navigate a dense IP and architecture landscape shaped by FERZ's filings, publications, and standards. Full portfolio details, claims maps, and prosecution stages are available under NDA.
IP Counsel: Hunton Andrews Kurth LLP and Resolute Legal PLLC
If deterministic assurance becomes a regulatory or operational requirement, the value accrues disproportionately to the architectural layer that sits between AI systems and consequential actions. The question for investors is not whether AI adoption grows. The question is where accountability ultimately settles.
Between now and that settlement, FERZ's position compounds. Patent applications advance through prosecution, defensive publications accumulate prior art, the published standard matures toward regulatory citation, and the research corpus expands. Each step increases the cost of competing entry and the value of the IP position FERZ already holds.
Position A: Now
Enter at the paradigm's origin, before the category formalizes. Invest in the company that published the theory, filed the patent applications, and is building the infrastructure, while the IP position remains available at seed-stage terms.
Position B: Later
Wait for category-forming regulatory or enforcement events to clarify the requirement, by which time the architectural pathways and the IP positions covering them will be settled, and the entry price will reflect what is then visible to every investor.
Investor Brief
The Market Timing Is Uncertain. The Control Requirement Is Not.
The full investment argument in six pages: why effect-bearing AI creates a structural requirement for pre-execution authorization, what is structurally required once that threshold is crossed, and why FERZ is commercializing the category it formalized.
FERZ Investor Brief
PDF · 6 pages · July 2026
Covers the runtime authorization boundary, the autonomy trilemma, the Five Tests Standard, and the FERZ investment thesis. Suitable for review ahead of a first conversation. The investor deck and terms are available on request; technical and architectural detail is available under NDA.
Not primarily. FERZ is an infrastructure and IP company. Software delivery may take SaaS, private deployment, OEM, or embedded form where appropriate, but the company should not be evaluated first as seat-based application SaaS. The primary assets are patent-pending enforcement architectures, a published standard, defensive publications, and a research corpus supporting the category.
Why are conventional SaaS metrics incomplete for FERZ?
Conventional SaaS metrics evaluate the efficiency of a distribution machine: recurring subscription growth, retention, acquisition payback, and seat expansion. FERZ is earlier in the value chain. Its position concentrates in the architectural layer between AI systems and consequential actions. At this stage, the relevant evaluation axes are coverage, defensibility, licensing leverage, regulatory timing, and control of the emerging category.
How does FERZ generate revenue?
FERZ is structured for infrastructure licensing, including portfolio licensing, field-of-use exclusives, OEM embedding, and deployment of runtime authorization infrastructure in regulated environments. Commercial terms and field-specific structures are discussed under NDA.
Why publish the Five Tests Standard instead of keeping everything proprietary?
A published standard defines the evaluation criteria for verifiable AI authorization. The Five Tests Standard (5TS) is vendor-neutral so implementations can be assessed against a common framework. FERZ's position comes from building the architecture against those criteria from origin, while maintaining patent filings, defensive publications, and trade-secret protection around the broader portfolio.
Founding Team
Founding Team
Edward Meyman
Founder and CEO
Theory, IP, Doctrine
Created the theoretical and IP foundation for deterministic assurance. Inventor on all patent filings. Author of the published research corpus that defines the category. Designed the dual-layer IP strategy and enforcement architecture.
Eugene Saksonov
Co-Founder and CTO
Engineering, Architecture
Translates formal assurance theory into production-grade enforcement systems. Owns the engineering pathway from mathematical specification to deployable infrastructure. Final authority on all technical claims.
Neil Nair
Co-Founder and CBO
IR, GTM, Regulatory
Drives investor relations and enterprise go-to-market across healthcare, financial services, and government. Leads regulatory engagement and standards-body positioning. Primary point of contact for investment conversations.
Request Investor Briefing
Request Investor Briefing
Confidential Investor Intake
All requests are reviewed directly by the founding team.
FERZ is typically of greatest interest to investors focused on infrastructure, deep technology, regulated industries, and category-forming platforms. Engagement begins with a direct briefing with the founding team. A structured evaluation package, published research, IP portfolio overview, and architectural thesis are provided as appropriate during diligence.