Disclaimer
Last updated: June 2026
Important Notice
EVE AI Core is an autonomous governance architecture for AI systems. It implements deterministic charter enforcement, cryptographic audit trails, and zero-trust identity invariants. All cognitive and affective computations are algorithmic simulations operating within defined safety boundaries.
Nature of the Platform
EVE AI Core is a deterministic governance enforcement platform for AI systems. The architecture includes:
- Three-Plane Architecture (Control, Execution, Evidence) with hardware isolation
- TVE (Telemetry Verification Engine) 8-stage pipeline for output verification
- CRD (Confidence-Reality Divergence) scoring with domain-specific calibration
- Deterministic charter enforcement with 15 immutable rules and 12 principles
- Cryptographic audit trail via SHA-256 hash chains and HMAC-SHA256 attestations
- Truth Store with pre-computed reality verification and contradiction detection
- Circuit breaker with three-state machine for cascade failure prevention
- AEGIS automated adversarial testing with 210+ danger detection patterns
Governance decisions are deterministic and reproducible. Verification scores, veto verdicts, and attestation signatures represent computational outputs of the enforcement pipeline, not subjective assessments.
Governance Architecture
EVE AI Core enforces behavioral constraints through deterministic governance systems:
- Charter Override Protection — 12 immutable principles enforced at the architecture level before any action is dispatched
- Claims Ledger — Cryptographic hash-chain provenance for all assertions with Brier score calibration
- Unified Trust Dial — Domain-specific confidence thresholds gating assertion levels
- Action Registry — Three-phase propose/approve/execute workflow for all system actions
- Self-Modification Governance — Configuration drift budgets, capability escalation review, and human approval chains for high-risk changes
- Reality Anchor Agent — Continuous belief grounding against verified baselines
Research and Operational Purpose
This platform is designed for:
- Applied research in autonomous AI governance architectures
- Development and testing of charter-based ethics enforcement systems
- Exploration of cognitive architecture patterns for AI safety
- Production deployment of integrity-verified AI governance systems
- Multi-agent coordination and trust framework research
Governance Limitations
While EVE AI Core implements deterministic governance enforcement, the following limitations apply:
- Governance enforcement operates on AI outputs — it does not control the underlying model's internal reasoning
- CRD scores are calibrated against verified facts but cannot guarantee detection of all novel misinformation
- The system enforces policy rules deterministically, but the quality of governance depends on the rules defined
- Hardware-rooted enforcement (AIMS / PolarFire SoC) is on the roadmap for Sovereign-tier deployments and is not yet generally available
- Cryptographic attestations prove that governance was applied, not that the governed output is factually correct
What EVE AI Core Is
- A deterministic governance enforcement layer that sits between AI systems and production
- A Three-Plane Architecture: Control (decisions), Execution (enforcement), Evidence (proof)
- A CRD scoring engine that measures confidence-reality divergence on every AI output
- A cryptographic audit trail producing HMAC-SHA256 signed attestations on every governance decision
- An enterprise infrastructure product licensed at $150K–$1.2M+ annually
What EVE AI Core Is Not
- A replacement for human oversight in safety-critical decisions
- A guarantee that governed AI outputs are factually correct — it enforces policy, not truth
- A substitute for regulatory compliance programs — it provides infrastructure for compliance, not compliance itself
- Infallible — the system includes hallucination detection and AEGIS adversarial testing precisely because no system is perfect
- A chatbot, virtual assistant, or consumer AI product
Data Processing and Privacy
By using EVE AI Core, you acknowledge that:
- Conversations are processed and may be stored in multi-layer memory systems for service improvement
- Memory governance controls allow users to set retention policies, request deletion, and protect specific memories
- Cross-user pattern learning is strictly opt-in with privacy-preserving aggregation (hashed user IDs, minimum contribution thresholds)
- All data processing is subject to our Privacy Policy
- Users should not share sensitive credentials, financial information, or classified data through the chat interface
- The immutable audit ledger maintains a cryptographic record of system actions for accountability
Support Services
EVE CoreGuard Support is provided on a commercially reasonable, best-effort basis according to the service-level targets associated with your subscription plan, the category of your request, and its assigned priority. You acknowledge that:
- Stated first-response and resolution targets are objectives, not guarantees, and may be affected by factors outside our reasonable control
- Support is not an emergency or life-safety service and must not be used to report situations requiring urgent intervention
- Support records are isolated to your organization and recorded in a tamper-evident audit trail; where you reference a governance Decision ID, the related signed evidence may be retrieved to investigate your request
- Files you attach must conform to the permitted file types and size limits and must not contain malicious or unlawful content
No Warranty
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. THE GOVERNANCE SYSTEMS, WHILE DESIGNED FOR RELIABILITY, DO NOT GUARANTEE PREVENTION OF ALL UNDESIRED OUTPUTS.
Limitation of Liability
IN NO EVENT SHALL THE AUTHORS, COPYRIGHT HOLDERS, OR EVE AI CORE BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. THIS INCLUDES BUT IS NOT LIMITED TO DAMAGES ARISING FROM RELIANCE ON AI-GENERATED CONTENT, THERAPEUTIC INTERACTIONS, OR GOVERNANCE SYSTEM DECISIONS.
Governance Integrity Statement
EVE AI Core's governance architecture is designed with the following guarantees:
- Charter principles are immutable and cannot be overridden by user requests, system modifications, or social engineering
- All system actions pass through deterministic compliance checks before execution
- Self-modification proposals are subject to risk assessment, charter compliance verification, and configuration drift budget enforcement
- A complete audit trail is maintained with cryptographic integrity verification
These guarantees operate at the architectural level and are independent of LLM behavior.
Intellectual Property
EVE AI Core and its associated governance architecture, cognitive systems, and documentation are the intellectual property of Jamaurice Holt. Patent applications are pending for multiple novel systems within the platform. Unauthorized reproduction, distribution, or derivative use is prohibited.
Contact
For legal inquiries about this disclaimer, email legal@eveaicore.com. For questions about the governance architecture or EVE AI Core generally, see our About page or the Secure Communications form.