Definition Document

Runtime Immunity

Structure-Based Runtime Protection Framework

v1.1 · Fully Integrated Edition · March 2025

Runtime Immunity Definition Document v1.1

Essential Definition and Technical Framework of Runtime Immunity

v1.1 (Fully Integrated Edition) · March 2025 · Public Edition


§1Conceptual Definition

§1.1Single-Sentence Definition of Runtime Immunity

Runtime Immunity: A technology framework that structurally nullifies the preconditions or outcomes of attacks even when attacks reach the system, maintaining protection attributes without halting the system.

§1.2Expanded Definition: Structure-Based Protection

"Severing the Causal Relationship Between Breach and Damage"

Conventional security has implicitly assumed that "defense breach = damage occurrence." Runtime Immunity is a technology framework that structurally severs this causal relationship itself. Even when an attack "reaches" the system, the "outcome" holds no usable value.

Why "Structural Nullification"?

Because it does not depend on detection, it can address unknown threats (0-day). Because it requires no human response, it provides consistent 24/7 protection. Furthermore, by structurally severing the causal relationship between attack success and damage occurrence, it structurally minimizes legal and financial risk (damages claims, fines, GDPR penalties) and structurally makes it difficult for breach notification obligations to arise.

⚠️ NOTICE: Regarding "Structurally Infeasible": "Structurally infeasible" does not mean "100% impossible." Claims of completeness may become a legal vulnerability; therefore, this definition adopts expressions such as "structural minimization" and "computationally infeasible."

§1.3Two Design Principles of Runtime Immunity

Principle 1: Detection Independence

The effectiveness of protection mechanisms does not depend on the success or failure of attack detection. Even when detection gaps exist, attack outcomes are structurally nullified. This is the fundamental difference from Runtime Security and the greatest strength of Runtime Immunity.

Principle 2: Structural Embedding

Protection is embedded into the system architecture and does not depend on runtime detection or decision-making. This includes protections implemented at the design stage, such as memory encryption, process isolation, and environment binding.

§1.4Characteristics Derived from Runtime Stability Design Philosophy

Characteristic 1: Non-Halting

The system is not halted even under attack. This is an essential characteristic for protecting "unstoppable systems" such as autonomous vehicles, medical devices, and industrial control systems.

Characteristic 2: Homeostasis Maintenance

The system's protection attributes remain unchanged before and after an attack. Similar to biological homeostasis, the system autonomously maintains a stable state internally.

§1.5Position within Runtime Stability and Contribution to 7 Attributes

Runtime Immunity is positioned as structure-based protection within the Runtime Stability framework. It makes a direct contribution particularly to Inexploitability.

Table 1: Runtime Immunity's Contribution to the 7 Attributes

AttributeContributionRuntime Immunity's RoleRelationship with Runtime Security
Safety★★★★★Structurally prevents transitions to dangerous statesCoordinates with monitoring and detection
Reliability★★★★★Guarantees automatic continuation of protection functionsCoordinates with monitoring and recovery
Availability★★★☆☆Somewhat limited structurallyPrimary responsibility of Security
Controllability★★★★☆State awareness and steering mechanismsCoordinates with dynamic control
Confidentiality★★★★★Fundamental protection through encryptionCoordinates with detection and restriction
Data Integrity★★★★★Implements tamper-proof structuresCoordinates with detection and prevention
Inexploitability★★★★★Structural nullification of attack outcomesLimited in Security alone

§2Problems Addressed

§2.1Limitations of Detection-Dependent Security

Detection-based security such as Runtime Security has inherent limitations. Detection gaps inevitably exist, unknown threats (0-day) cannot be addressed, and temporal delays allow damage to occur.

Runtime Immunity fundamentally resolves these limitations by structurally nullifying attack outcomes regardless of detection success or failure.


§3Technical Definition

§3.1Realization Structure of Runtime Immunity

Runtime Immunity nullifies attack outcomes through three realization forms corresponding to each stage of the attack chain.

§3.2Three Realization Forms

Form 1: Target Elimination

Achieves a state where attackers cannot identify or locate attack targets. Example: In an environment where critical server locations are concealed on the network and all process memory is uniformly encrypted, attackers cannot locate valuable data.

Form 2: Precondition Removal

Achieves a state where the "tools" and "conditions" required for an attack are structurally absent. Example: Encryption keys are not persistently stored but dynamically derived from the physical environment only at execution time. Access privileges required by malware are structurally eliminated.

Form 3: Spoils Nullification

Even when an attack reaches and executes, the obtained outcomes hold no usable value. Example: A successful memory dump yields only undecryptable ciphertext. Stolen authentication credentials are bound to the execution environment and are unusable in other environments.


§4Nullification Level Definition

§4.1Why Level Definitions Are Necessary

The degree of "structural nullification" varies by implementation. Since claiming completeness poses legal risk, nullification is defined quantitatively.

§4.2Two-Axis Nullification Evaluation Framework

Key Recognition: The information-theoretic axis and economic rationality axis are independent axes.

Axis A: Information-Theoretic Nullification

Mathematical decryption infeasibility based on Shannon's theorem. Evaluation based on classical computer computational capabilities.

NL-1 (Partial Information-Theoretic Nullification)

  • Condition: 50% or more of data obtainable through attack is computationally infeasible to decrypt
  • Example: 50% of process memory encrypted with security parameter of 128/256 bits or higher
  • Risk: Partial encryption means potential information leakage from unencrypted regions

NL-2 (Comprehensive Information-Theoretic Nullification)

  • Condition: 95% or more of data obtainable through attack is computationally infeasible to decrypt
  • Example: All process memory encrypted with security parameter of 128/256 bits or higher, keys not stored
  • Assurance: Computationally infeasible to decrypt as long as implementation conditions are met

Axis B: Economic Nullification

A state where "investment cost > obtainable benefit" is mathematically established from the attacker's perspective.

NL-3 (Practical Nullification)

  • Prerequisite: NL-2 (comprehensive encryption) must be implemented
  • Additional implementation: Per-session environment binding, one-time tokenization of credentials, privilege minimization
  • ROI analysis: Even if decryption succeeds, attacker's practical benefit is eliminated
  • Assessment: Theoretically the strongest class

§4.3Preparation for Post-Quantum Nullification (PQN)

Premises of current NL definitions:

  • Shannon's theorem (classical cryptography)
  • Security parameter of 128/256 bits or higher

Quantum-era threats:

  • Practical quantum computers: Expected 2030s–2040s
  • Shor's algorithm: Potential to break RSA, ECC, etc.

⚠️ NOTICE: Position of this document (v1.1): Recorded as a definition for the classical cryptography era. Valid as implementation criteria for 2025–2030. Updates are required for the post-quantum era after 2030.


§5Achievement Evaluation Framework

§5.1Runtime Immunity Level (IL-0 to IL-3) and Nullification Levels

LevelDescriptionNL Correspondence
IL-0No structural protectionNo structural protection in runtime environment
IL-1NL-1 (partial nullification) achievedPartial implementation of a single form
IL-2NL-2 (comprehensive nullification) achievedIntegrated implementation of multiple forms (2+)
IL-3NL-3 (practical nullification) achievedComprehensive implementation of all three forms

§6Relationship with Existing Technologies

§6.1Relationship with Confidential Computing (TEE, SGX, etc.)

Runtime Immunity is a technology framework that may include but is not limited to Confidential Computing. TEE and SGX are excellent realization means for Runtime Immunity, but they have the limitation of hardware dependency.


§7Application Domains

Runtime Immunity applies to "unstoppable systems," "systems that cannot tolerate damage," and "systems requiring comprehensive protection."

  • Autonomous vehicles and connected cars
  • Medical devices and healthcare information systems
  • Core business systems in financial institutions
  • Industrial control systems and smart grids
  • Critical social infrastructure
  • Defense-related systems

§8Glossary

  • Nullification: A state where data obtained through attack holds no usable value for the attacker
  • Information-Theoretic Nullification: An encryption state that is computationally infeasible to decrypt
  • Economic Nullification: A state where the attacker's ROI is negative even if decryption succeeds
  • Post-Quantum Cryptography (PQC): Cryptographic technology resistant to quantum computers
  • Environment Binding: Technology that dynamically binds authentication information to the execution environment

§9Revision History

  • v1.0 (March 2025): Initial release
  • v1.1 (March 2025) Precision Edition: Explicit separation of NL axes. Added PQN preamble.
  • v1.1 (March 2025) Fully Integrated Edition: Integrated v1.1 revisions into the complete v1.0 body.

§10About This Document

§10.1Purpose

This definition document establishes the essential definition of Runtime Immunity and systematically positions the technology framework for structurally nullifying attack outcomes within the Runtime Stability framework.

§10.2Cross-References

  • Runtime Stability Definition v3.3: Overall framework concepts and 7-attribute integration
  • Runtime Security Definition v1.1: Detection-based protection system (3-axis SL evaluation, including RS-6 limitations)

*Runtime Stability — Technology that controls an unstoppable world. Breached, yet unshaken.*