Peter Lablans wins patent for new cryptographic computational functions

Researcher Peter Lablans has secured U.S. Patent 12,665,774 for a cryptographic approach that replaces familiar modular arithmetic with a new class of machine-level functions. The invention is designed to expand the design space for encryption, hashing and authentication as cryptography faces pressure from quantum computing and AI.

Why it matters: - The patented framework could expand cryptography beyond the small set of primitives that have defined the field for decades. - The approach is designed to create new options for encryption, hashing and authentication without abandoning existing cryptographic architectures. - The invention is aimed at future systems that need to adapt to threats from quantum computing and artificial intelligence.

What happened: - Peter Lablans was granted U.S. Patent 12,665,774 for a new class of cryptographic computational functions. - The patent covers a machine-level operation that introduces non-reversible diffusion directly into computation while preserving deterministic behavior and statistical balance. - The work centers on Morristown, New Jersey, where the patent announcement was issued June 25, 2026.

The details: - The patented framework operates on non-binary elements rather than conventional binary arithmetic. - At 256 states based on 8-bit bytes, the number of possible two-operand functions exceeds 10^150,000. - In this model, keys, nonces and plaintext remain variables, while functions are treated as immutable constants in conventional cryptography. - The invention changes a function into a variable, creating a polymorphic implementation. - The operation combines a reversible function with a randomized non-reversible transition function to create a deterministic nonlinear mixing computer operation. - The patent says the transformation preserves the computational architecture of existing designs while expanding the family of machine-level functions or implementations that can realize them. - The stated characteristics include high nonlinearity and diffusion while maintaining statistical neutrality. - The framework is designed to remain compatible with existing cryptographic architectures. - The invention is intended to enable vast computational diversity and more agile implementation. - A detailed article with numerical examples is available here. - Peter Lablans specializes in computational logic, cryptographic primitives and computer architecture.

Between the lines: - Lablans is arguing that cryptography has focused too narrowly on algorithms built from the same underlying ingredients. - The core claim is that security has been studied mostly at the level of mathematical primitives, while the space of possible implementations has been underused. - The reference to Dr. Gerrit Blaauw and IBM System/360 frames the invention as an extension of architecture-versus-implementation thinking into cryptographic functions themselves. - The pitch is as much about expanding the search space for cryptographic design as it is about a single new operation.

What's next: - Lablans says the research points toward cryptographic systems that can evolve as threats change. - The patent and accompanying article are positioned as a foundation for further work in machine cryptography. - The broader test will be whether the approach can be translated into practical encryption, hashing and authentication systems.

The bottom line: - The patent claims a new computational landscape for cryptography, not just a new algorithm.