AI summary by TwinMind
Quantum Compute with Single Photons
Asianometry
24m 3s
Classical vs Quantum Computing Fundamentals
- Classical computers: store info in bits (0/1), manipulate via gates (NOT, AND, NAND, OR, NOR)
- Nonlinear gates essential for complex calculations - electrons interact strongly
- Quantum computers: store info in qubits with superposition (0 and 1 simultaneously)
- Quantum gates: Pauli X (like NOT), Hadamard (creates 50/50 superposition), CNOT (controlled NOT)
- Entanglement: linking qubits so measuring one instantly affects the other
- Necessary for useful algorithms like Shor's factoring (breaks RSA encryption)
Matter-Based Quantum Computing Approaches
- Trapped ion: atoms suspended via electromagnetic forces, manipulated with lasers
- Superconducting circuits (IBM/Google): Josephson junctions manipulated with microwaves
- Google's Willow chip: 105 qubits (Dec 2024), previous chip had 72 qubits
- Stock rose 6% after announcement
- Need ~1 million qubits for useful algorithms
- Challenges: decoherence from environment (cosmic rays, thermal noise)
- Require extreme cooling/vacuum systems
- Quantum error correction adds complexity and potential errors
Photonic Quantum Computing Advantages
- Photons: massless, travel at light speed, don't interact with environment easily
- Lower decoherence risk, potential room temperature operation
- Can scale to million qubits more easily
- Challenge: photons don't interact with each other naturally
- Cannot create nonlinear gates needed for entanglement
KLM Scheme Breakthrough (2001)
- Knill-Laflamme-Milburn paper in Nature
- Achieves nonlinear results from linear optical equipment
- Data qubits: single photons encoded via polarization or dual-rail paths
- Ancilla photons: helper photons with known states
- Process: mix data/ancilla photons through beam splitters/phase shifters
- Measure ancilla photons (not data qubits)
- If specific outcome detected, quantum gate successfully applied
- ~25% success rate for CNOT gates
- Problem: probabilistic nature makes circuits impractical (25% × 25% = 6.25% for two gates)
Measurement-Based Computation
- Shift from building circuits to using measurement as computation tool
- Create standardized entangled "resource states" (cluster states)
- SiQure uses 6-photon ring formations
- Fusion network: merge resource states via measurement to create massive entanglement lattice
- Can work at own pace (not light speed)
- Final network becomes the calculation
Technical Implementation Challenges
- Single photon generation: need identical photons
- Hong-Ou-Mandel effect: two identical photons hitting 50/50 beam splitter exit together
- Enables quantum entanglement between photons
- Detection: superconducting nanowire single photon detectors (SNSPDs)
- Ultra-thin wires with supercurrent
- Single photon breaks Cooper pairs, creates voltage spike
SiQure Company Details
- Founded 2015, pursuing photonic quantum computing
- Published Nature paper on manufacturable platform using Global Foundries silicon photonics
- Developing quantum campuses in Illinois and Australia
- Goal: utility-scale quantum compute
Competitive Landscape
- Fela (Europe): targeting 50 logical qubits by 2028
- Jiuzhang (China): University of Science and Technology
- Three versions 2020-2023
- Gaussian boson sampling in 200 seconds vs 2.5 billion years for classical supercomputer
- Limited practical application currently
Current Limitations and Future Outlook
- All encryption remains intact
- Quantum computing won't replace classical computers
- Different use cases: high compute, little memory
- Applications: cryptography, weather prediction, drug simulation
- LLMs like ChatGPT not suitable (memory requirements)
- Goal: bring quantum compute to developers for experimentation
