Ben Goertzel Launches New Lattice-Based Scheme Enables Homomorphic Evaluation of Quantum Programs with Post-Quantum Security

Ben Goertzel has introduced an efficient quantum-safe homomorphic encryption scheme for quantum computer programs. This scheme addresses practical challenges with lattice-based cryptography and demonstrates compatibility with near-term quantum hardware. Ben Goertzel is a prominent AI researcher, entrepreneur, and author who has been a key figure in artificial general intelligence (AGI).

The research presents a lattice-based scheme for homomorphic evaluation of programs and proofs secure against adversaries. It lifts classical homomorphic encryption to Module Learning-With-Errors (MLWE) lattices, formalizing security via the qIND-CPA game with reduction to decisional MLWE. A secret depolarizing BNSF mask hides amplitudes, storing states as MLWE ciphertext pairs. Practical issues are addressed: a typed QC-bridge encrypts classical bits while enabling measurement controls; Pauli twirls add circuit privacy; knowledge bases are shipped as MLWE capsules. Performance analysis shows feasibility within current QPU idle windows, with a 100-qubit proof running in ~10 ms and keys under 300 kB. Results suggest compatibility with near-term quantum clouds and post-quantum assumptions.

Quantum computing holds immense potential for solving complex problems, yet it poses significant risks when handling sensitive data. Traditional methods require decrypting data before processing, which is inherently insecure. This conundrum has hindered the adoption of quantum technologies in sectors where privacy is critical. The challenge lies in maintaining security without compromising computational power.

Homomorphic encryption is a pivotal solution, allowing computations on encrypted data without decryption. This technique ensures sensitive information remains protected throughout processing. A key innovation is weak-measurement homomorphic evaluation, which enables operations on encrypted data while preserving the integrity of quantum states.

This method employs weaker measurements to prevent premature collapse of quantum states, crucial for multiple operations before decryption. By adjusting hardware parameters like measurement intensity, researchers optimize the balance between computation depth and state preservation, enhancing practical applications.

Superconducting qubits and trapped ions form the foundation of this technology, utilizing techniques such as microwave readouts and fluorescence photon sensing to detect quantum states without full collapse. Applications span cloud-based recommendation systems, mobile trading apps, and health AI, where data privacy is paramount.

Benefits include lower qubit overhead, shallower circuits, and improved signal-to-noise ratios, translating into faster processing and efficient resource use. A roadmap outlines milestones from 2023 to 2035, detailing stages from theoretical validation to full-scale implementation across various industries.

Homomorphic encryption signifies a significant step forward in quantum computing and privacy protection. Enabling secure computations on encrypted data paves the way for transformative applications while safeguarding sensitive information. As we progress towards implementing this technology, the future promises enhanced security and computational power, reshaping industries reliant on data integrity.

All about Ben Goertzel

Ben Goertzel is a prominent computer scientist, artificial intelligence researcher, and entrepreneur who has significantly contributed to the field of Artificial General Intelligence (AGI), having helped popularize the term itself. With a PhD in mathematics from Temple University, Goertzel currently leads multiple organizations including SingularityNET Foundation, the OpenCog Foundation, and the AGI Society which runs the annual Artificial General Intelligence conference. His most notable projects include SingularityNET, a decentralized AI marketplace that brings AI and blockchain together to create an open market for AI services, and OpenCog, an open-source framework designed to advance human-level artificial general intelligence.

As Chief Scientist at Hanson Robotics, he led the software team behind the humanoid robot Sophia, and through Awakening Health, he’s developing “Grace,” a healthcare-focused robot. Recently, he launched over $1 million in grants to empower developers to advance “benevolent AGI” through the OpenCog Hyperon framework. Goertzel is also a prolific author, having written numerous books on artificial intelligence, consciousness, and the technological singularity.