IBM ResearchLab

Actively publishes on AI for materials property prediction including thermal conductivity of 2D materials and advanced packaging; cryogenic thermal management for quantum hardware adds urgency.

Long-standing nanoscience division (STM manipulation, 2D materials) combined with an active RL research group; positioned to fuse both in autonomous nano-assembly.

Pioneered phase-change and memristive synaptic devices for analog AI inference; publishes actively on biophysically-motivated hardware for in-memory computing.

Decades of analog AI hardware work using phase-change memory and published spintronic AI inference research places them at the precise bridge of Field A and Field B.

RoboRXN and broader AI-for-materials programs target automated discovery loops coupling simulation, synthesis, and learning — aligns with thermoelectric and battery bridge fields.

Operates both TrueNorth neuromorphic and superconducting qubit programs; has in-house cryogenic electronics expertise and chip fabrication that could bridge the two.

Active in both neuromorphic computing architectures and AI-accelerated materials simulation; thermal-aware chip design is a documented research theme within their hardware research division.

Long-running materials-informatics program combining atomic-scale simulation, physical characterization, and machine learning for nanomaterial and device discovery.

Pioneered PCM-based neuromorphic synapses AND operates IBM Quantum; phase-change materials are the hardware bridge between the two fields, and IBM holds depth on both sides.

Active in AI-assisted quantum circuit compilation and transpilation (Qiskit ecosystem) and ML-guided noise characterization across Eagle/Heron processor generations.

IBM's analog AI hardware team uses phase-change memory and resistive elements operating at voltage-defined conductance states for in-memory matrix multiply

Leads industrial Josephson junction nanofabrication for superconducting processors; actively publishing on junction scaling, materials, and cryogenic integration.

Has published on 2D-material photocatalysis and photoelectrochemical systems, with materials-physics depth that maps directly onto the TMD and band-gap bridge fields identified here.

IBM's photonic computing and neuromorphic (NorthPole, analog AI) programmes explicitly grapple with phonon-photon thermal crosstalk in integrated platforms.

Has published on AI for materials property prediction and is motivated by chip-level thermal management; sits at the intersection of both fields institutionally.

Has published seminal work on PCM and analog crossbar arrays for AI inference, directly matching the representative B paper lineage; multiple deployed prototype chips.

Pioneered phase-change memory (PCM) synaptic devices—electrochemical electrode-based—for in-memory neuromorphic AI; directly at the intersection of both fields.

Long-standing programs in both nanophotonics and neuromorphic computing; published on analog AI hardware and optical computing in adjacent spaces.

Runs the world's largest superconducting qubit program and faces acute thermal-management and decoherence challenges that require deep phonon-level understanding of heat flow in superconducting circuits.

Runs end-to-end superconducting quantum processor development including Josephson junction fabrication, where thin-film chemical deposition quality is a key performance bottleneck.