top of page

Inside AMD and PsiQuantum’s $1 Billion-Scale Ambition to Build the Next Generation of Computing

1 day ago
11 min read
Quantum computing is approaching a critical engineering phase. After years of research focused on demonstrating quantum bits, improving coherence, and developing experimental processors, the industry is increasingly confronting a more demanding challenge: building complete systems capable of performing useful computations reliably at commercial scale.

A quantum processor cannot operate independently of the classical infrastructure surrounding it. It requires conventional processors, specialized control electronics, high-speed data processing, error-correction systems, and software capable of coordinating complex operations. As quantum machines become larger, these supporting technologies become essential to their performance.

The expanded partnership between semiconductor manufacturer AMD and photonic quantum computing developer PsiQuantum reflects this shift. Announced on November 12, 2026, the collaboration includes a planned equity investment by AMD and a joint effort to design the classical and quantum computing infrastructure required for utility-scale quantum systems.

The partnership combines AMD's CPUs, GPUs, adaptive computing technology, and software ecosystem with PsiQuantum's silicon photonics platform. Its objective extends beyond connecting two technology companies. It addresses one of the central engineering problems in quantum computing: how to coordinate large-scale classical processing with quantum hardware while maintaining the speed and precision required for fault-tolerant computation.

The collaboration also has an immediate commercial dimension. Both companies are applying high-performance computing to computational chemistry and oncology drug discovery while preparing infrastructure for future quantum applications.

This combination of near-term classical computing benefits and long-term quantum development illustrates how the industry may progress toward practical quantum systems.

Why Quantum Computing Needs Powerful Classical Infrastructure

Quantum computers use quantum bits, or qubits, to represent and manipulate information. Unlike conventional binary bits, which represent either zero or one, qubits can occupy quantum states involving superposition and entanglement.

These properties allow quantum algorithms to approach certain computational problems differently from classical algorithms. Potential applications include molecular simulation, materials research, optimization, and selected problems in cryptography and scientific computing.

However, quantum processors are extremely sensitive systems. Environmental disturbances, imperfect operations, and hardware limitations can introduce errors into calculations. Consequently, a useful large-scale quantum computer must do much more than execute quantum gates. It must continuously monitor and manage the reliability of its operations.

This is where classical computing becomes indispensable.

A fault-tolerant quantum computer requires conventional hardware to process measurement results, identify error patterns, calculate corrections, and coordinate control operations. These activities must happen within tight timing constraints.

In a photonic quantum architecture, information is carried by particles of light, or photons. The system must coordinate optical components, detectors, control electronics, and classical processors. As the number of quantum operations increases, the supporting computational workload can become substantial.

The relationship can be understood through three interconnected layers:

Quantum processing: Performs operations using qubits and quantum circuits.

Classical control and error correction: Interprets measurement outcomes, detects errors, and coordinates corrections.

High-performance computing and software: Provides simulation, algorithm development, workload orchestration, and computational support.

A weakness in any layer can limit the performance of the complete system. Faster quantum hardware alone cannot guarantee useful results if classical processing cannot keep pace.

AMD and PsiQuantum are therefore addressing a system-level engineering problem rather than simply combining processors from two manufacturers.

AMD's Strategic Investment in PsiQuantum

AMD's planned equity investment creates a closer financial relationship between the semiconductor company and PsiQuantum. The companies have not disclosed the investment amount, but the announcement establishes a broader partnership focused on developing infrastructure for utility-scale quantum computing.

The agreement brings together complementary technologies.

AMD develops CPUs, GPUs, adaptive computing devices, and software platforms used across data centers, artificial intelligence, and high-performance computing. PsiQuantum is developing a photonic quantum computing architecture based on silicon photonics.

Under the expanded arrangement, AMD will contribute:

EPYC central processing units for general-purpose computing.

Instinct graphics processing units for accelerated workloads and large-scale simulation.

Versal adaptive system-on-chips for programmable computing and real-time control.

ROCm software for GPU programming and workload orchestration.

PsiQuantum will contribute its silicon photonics technology, quantum system architecture, and software and computational workflows.

The investment is strategically significant because it aligns the interests of a major semiconductor supplier with a company attempting to develop a complete quantum computing platform.

For AMD, the partnership offers an opportunity to extend its role into the emerging quantum infrastructure market. For PsiQuantum, access to AMD's computing technologies and engineering capabilities can help address the classical processing requirements of its architecture.

The collaboration also reflects a broader industry trend: quantum computing development increasingly depends on partnerships between specialized quantum hardware companies and established providers of classical computing infrastructure.

PsiQuantum's Silicon Photonics Approach to Quantum Computing

PsiQuantum is pursuing photonic quantum computing, an approach that uses photons as carriers of quantum information.

Photonic systems differ from architectures that use superconducting circuits, trapped ions, or neutral atoms. Instead of relying primarily on electrical or atomic quantum states, photonic quantum computers manipulate light through optical components and integrated photonic circuits.

One attraction of this approach is its potential compatibility with semiconductor manufacturing techniques.

Silicon photonics integrates optical functions with semiconductor-based technology. It allows light to be generated, routed, manipulated, and detected using increasingly sophisticated photonic components.

This creates opportunities to manufacture complex optical systems using techniques related to established chip production.

A photonic quantum computer requires more than an optical processor. It needs a network of components capable of managing photons, routing signals, performing measurements, and coordinating quantum operations.

The architecture must also address practical challenges such as photon loss, detector performance, optical switching, and the reliability of large-scale component integration.

PsiQuantum's approach is built around the ambition of scaling photonic systems toward fault-tolerant quantum computing. Achieving that objective requires coordinating large numbers of optical components with classical computing systems that can respond rapidly to the quantum processor's operations.

AMD's adaptive computing technology and high-performance processors are intended to support this coordination.

Real-Time Quantum Error Correction and the PRISM ML Decoder

Error correction is among the most important requirements for building useful quantum computers.

Quantum information is fragile. Interactions with the surrounding environment and imperfections in hardware operations can introduce errors that compromise calculations.

Classical computers also experience errors, but they can generally use established digital techniques to detect and correct them. Quantum systems face a more complicated problem because directly measuring a qubit can disturb the quantum state being used for computation.

Quantum error-correction techniques address this by encoding logical information across multiple physical qubits. Additional measurements, known as syndrome measurements, provide information about potential errors without directly measuring the encoded logical state.

Classical processors analyze these measurement results to infer what corrections are required.

This creates a demanding computational workflow. Error detection and decoding must occur quickly enough to support the quantum processor's operation. Delayed processing can introduce bottlenecks that limit the speed and scale of the system.

PsiQuantum's PRISM machine-learning decoder is designed to address this challenge.

According to the partnership details, the companies have worked together to adapt PRISM for execution on AMD Versal adaptive system-on-chips.

These devices combine programmable logic with processing capabilities, allowing developers to implement specialized computing functions rather than relying exclusively on general-purpose CPUs.

The objective is low-latency syndrome decoding, enabling classical hardware to process quantum measurement information within the timing requirements of photonic control systems.

The collaboration highlights an important distinction between conventional high-performance computing and real-time quantum control.

Large simulations can benefit from massive parallel processing across GPUs. Error correction, however, can require highly predictable response times and specialized hardware configurations.

The most effective quantum infrastructure will likely combine both approaches, using programmable hardware for time-sensitive control and larger computing systems for simulation, algorithm development, and broader computational workloads.

AMD Instinct GPUs and EPYC CPUs: Building the Quantum-Classical Computing Fabric

The expanded partnership incorporates AMD's broader computing ecosystem.

EPYC CPUs provide general-purpose processing capacity for managing software, coordinating workloads, and handling the conventional computing tasks surrounding quantum operations.

Instinct GPUs provide accelerated computing for highly parallel workloads. GPUs are particularly effective at performing large numbers of mathematical operations simultaneously, making them valuable for scientific simulation, machine learning, and computational chemistry.

The ROCm software stack provides a programming environment for GPU-accelerated applications and helps developers build workloads that can use AMD hardware.

Together, these technologies can support a hybrid computing environment in which classical and quantum processors perform complementary tasks.

Infrastructure component

Technology

Intended role

General-purpose processing

AMD EPYC CPUs

System management and classical computation

Accelerated computing

AMD Instinct GPUs

Simulation, scientific workloads, and computational chemistry

Adaptive hardware

AMD Versal SoCs

Programmable control and real-time error decoding

Software environment

AMD ROCm

GPU programming and workload orchestration

Quantum processing architecture

PsiQuantum silicon photonics

Photonic quantum information processing

Error correction

PRISM ML Decoder

Processing quantum error syndromes

The architecture matters because quantum computers are unlikely to replace classical computing systems.

Instead, they are expected to operate as specialized accelerators within broader computational environments. Classical systems will continue handling tasks for which conventional processors are more practical, while quantum processors may eventually provide advantages for selected computational problems.

The AMD-PsiQuantum partnership is designed around this hybrid model.

Drug Discovery and Computational Chemistry: The Immediate Application

While utility-scale quantum computing remains a longer-term engineering objective, the partnership also targets an application with immediate commercial relevance: pharmaceutical research.

Drug discovery requires scientists to understand how molecules interact, how chemical structures influence biological activity, and how candidate compounds behave under different conditions.

Computational chemistry helps researchers investigate these questions before committing to expensive and time-consuming laboratory experiments.

One area of interest involves covalent inhibitors, a class of molecules relevant to oncology drug discovery. These compounds can form chemical bonds with specific biological targets, potentially influencing the activity of proteins associated with disease.

Predicting molecular interactions accurately is important because small changes in molecular structure can influence binding behavior, selectivity, and potential biological effects.

The companies say PsiQuantum's drug research technology is already running on AMD hardware and can produce more accurate results than existing methods.

They also report initial breakthrough results in the simulation of covalent inhibitors through the combination of PsiQuantum's workflow and AMD's GPU expertise.

This is an important distinction in the partnership's commercial strategy. The near-term value does not depend entirely on the arrival of a fault-tolerant quantum computer.

High-performance classical computing can already support sophisticated molecular modeling and scientific simulation. These capabilities can improve existing research workflows while providing a foundation for future hybrid quantum-classical applications.

The broader opportunity is to develop computational tools that combine several approaches:

Classical simulation for established molecular modeling tasks.

GPU acceleration for computationally intensive calculations.

Machine learning for analyzing complex scientific data.

Quantum algorithms for selected problems when suitable hardware becomes available.

Such a strategy could help pharmaceutical researchers investigate molecular systems that are difficult to model using conventional methods alone.

However, improved computational predictions must ultimately be validated through experimental research. Faster simulations do not independently establish clinical effectiveness or guarantee successful drug development.

Quantum Computing and the Future of Oncology Research

The pharmaceutical industry is a particularly relevant potential market for quantum computing because molecular systems involve complex interactions that can be computationally expensive to model.

Classical computers use numerical approximations to simulate quantum mechanical behavior in molecules. These methods are powerful and widely used, but their computational requirements can increase significantly as molecular complexity grows.

Quantum computers could eventually offer new ways to represent and investigate certain molecular systems.

Potential research applications include:

Molecular energy calculations.

Chemical reaction modeling.

Protein and ligand interaction research.

Materials discovery for pharmaceutical applications.

Optimization of selected experimental and manufacturing processes.

These applications remain dependent on the development of sufficiently capable quantum hardware and algorithms.

The AMD-PsiQuantum collaboration therefore has a two-stage commercial logic. GPU-powered computational chemistry can support near-term research, while the development of quantum infrastructure prepares the platform for more advanced workloads in the future.

For pharmaceutical companies, this approach could provide a way to explore emerging computational methods without waiting for the complete realization of large-scale quantum computing.

Chicago's Emerging Role in Quantum Infrastructure Development

The partnership also includes a potential regional infrastructure dimension.

PsiQuantum is the anchor tenant for the Illinois Quantum and Microelectronics Park in Chicago. The two companies are evaluating a co-located presence at the site as part of their broader expansion plans.

Co-location could support closer collaboration between hardware engineers, quantum researchers, software developers, and infrastructure specialists.

Building a utility-scale quantum system requires expertise across several disciplines. Photonic engineering, semiconductor technology, control systems, high-performance computing, software development, and scientific applications must work together.

A dedicated quantum campus can provide an environment for developing and integrating these technologies.

The companies also planned to present further technical details about their joint roadmap at the SC26 supercomputing conference in Chicago, scheduled for November 15 to 20, 2026.

The proposed collaboration reflects the increasing overlap between quantum computing research and the established high-performance computing industry.

The Challenges of Building Fault-Tolerant Quantum Computers

Despite the technological promise, several obstacles remain before quantum computing can deliver broad commercial value.

Error correction at scale: Fault-tolerant computing requires managing errors across large numbers of physical qubits. The classical infrastructure must process information quickly and reliably.

Photon loss: Photonic quantum systems must minimize losses during photon generation, transmission, manipulation, and detection.

System integration: A complete quantum computer requires coordination between optical components, control electronics, classical processors, and software.

Manufacturing scalability: Semiconductor-compatible manufacturing may support photonic integration, but producing large, reliable systems remains a substantial engineering challenge.

Software development: Quantum hardware must be supported by algorithms, programming tools, error-correction methods, and workflows that allow researchers and businesses to use it effectively.

Commercial economics: Building and operating advanced quantum infrastructure requires significant capital. Commercial applications must generate enough value to justify these investments.

The AMD partnership addresses some of the classical computing and infrastructure requirements, but it does not eliminate the fundamental challenges of quantum hardware development.

The distinction between a promising architecture and a commercially useful fault-tolerant system remains essential when assessing the industry's progress.

What the AMD-PsiQuantum Partnership Means for the Future of Computing

The expanded partnership reflects a shift in the quantum computing industry toward integrated infrastructure.

Earlier stages of development often emphasized the quantum processor itself. As systems become more sophisticated, the surrounding computing environment is receiving greater attention.

This creates opportunities for established semiconductor companies, high-performance computing providers, software developers, and specialized quantum technology firms.

AMD's involvement demonstrates how conventional computing technologies could become essential components of future quantum systems.

The partnership also points toward a more integrated model of scientific computing, in which CPUs, GPUs, adaptive hardware, and quantum processors work together.

For the wider industry, several developments will be important to monitor:

Demonstrated improvements in real-time quantum error decoding.

Performance and scalability of the PRISM decoder on AMD adaptive hardware.

Practical results from hybrid quantum-classical workflows.

Independent validation of computational chemistry improvements.

Progress toward larger and more reliable photonic quantum systems.

Commercial adoption of the combined infrastructure.

Concrete deployment plans associated with the Illinois Quantum and Microelectronics Park.

These milestones will provide a clearer picture of whether the collaboration can translate its technical ambitions into scalable computing infrastructure.

Conclusion: The Quantum Computing Race Is Becoming an Infrastructure Race

AMD's planned equity investment in PsiQuantum and their expanded technology partnership highlight an important reality: the future of quantum computing will depend on much more than quantum processors.

Fault-tolerant systems require classical computing infrastructure capable of handling error correction, control operations, simulation, and software orchestration. AMD's CPUs, GPUs, Versal adaptive devices, and ROCm platform provide technologies intended to support these requirements, while PsiQuantum contributes its silicon photonics architecture and quantum computing expertise.

The partnership also creates opportunities in computational chemistry and oncology research, where GPU-powered workflows can deliver practical value before large-scale quantum hardware becomes commercially available.

For Dr. Shahid Masood and the expert team at 1950.ai, this development illustrates how the next generation of computing may emerge from the convergence of artificial intelligence, high-performance computing, semiconductor engineering, and quantum technology.

The central question is no longer simply how to build a more powerful quantum processor. It is how to construct an integrated system in which quantum hardware and classical infrastructure operate together reliably, efficiently, and at scale.

Key Takeaways

AMD announced a planned equity investment in PsiQuantum on November 12, 2026, as part of an expanded quantum infrastructure partnership.

The investment amount has not been disclosed.

AMD will contribute EPYC CPUs, Instinct GPUs, Versal adaptive computing technology, and the ROCm software stack.

PsiQuantum will contribute its silicon photonics platform and quantum computing expertise.

The companies are adapting the PRISM machine-learning decoder to AMD Versal hardware for low-latency quantum error correction.

The partnership combines quantum processing with classical high-performance computing and adaptive control systems.

Computational chemistry and oncology drug discovery are important near-term application areas.

The companies report initial results involving the simulation of covalent inhibitors.

Both organizations are evaluating a co-located presence at the Illinois Quantum and Microelectronics Park in Chicago.

The collaboration highlights the importance of classical computing infrastructure in developing fault-tolerant quantum computers.

Future Outlook

The AMD-PsiQuantum partnership represents a broader transition in quantum computing, from isolated processor demonstrations toward integrated systems designed for practical workloads.

The industry's progress will depend on advances in photonic hardware, error correction, classical processing, software integration, and commercial applications. GPU-accelerated scientific computing may provide valuable capabilities during this transition, while future fault-tolerant quantum systems could expand the range of computational problems researchers can investigate.

If these technologies mature together, the convergence of classical and quantum computing could become an important foundation for scientific discovery, pharmaceutical research, advanced materials, and next-generation computing infrastructure.

Further Reading / External References

AMD to invest in PsiQuantum to advance quantum computing infrastructure

https://www.reuters.com/business/amd-invest-psiquantum-advance-quantum-computing-infrastructure-2026-11-12/

PsiQuantum and AMD Expand Partnership with Equity Investment to Co-Design Utility-Scale Quantum Infrastructure

https://quantumcomputingreport.com/psiquantum-and-amd-expand-partnership-with-equity-investment-to-co-design-utility-scale-quantum-infrastructure/

Quantum computing is approaching a critical engineering phase. After years of research focused on demonstrating quantum bits, improving coherence, and developing experimental processors, the industry is increasingly confronting a more demanding challenge: building complete systems capable of performing useful computations reliably at commercial scale.

A quantum processor cannot operate independently of the classical infrastructure surrounding it. It requires conventional processors, specialized control electronics, high-speed data processing, error-correction systems, and software capable of coordinating complex operations. As quantum machines become larger, these supporting technologies become essential to their performance.


The expanded partnership between semiconductor manufacturer AMD and photonic quantum computing developer PsiQuantum reflects this shift. Announced on November 12, 2026, the collaboration includes a planned equity investment by AMD and a joint effort to design the classical and quantum computing infrastructure required for utility-scale quantum systems.

The partnership combines AMD's CPUs, GPUs, adaptive computing technology, and software ecosystem with PsiQuantum's silicon photonics platform. Its objective extends beyond connecting two technology companies. It addresses one of the central engineering problems in quantum computing: how to coordinate large-scale classical processing with quantum hardware while maintaining the speed and precision required for fault-tolerant computation.

The collaboration also has an immediate commercial dimension. Both companies are applying high-performance computing to computational chemistry and oncology drug discovery while preparing infrastructure for future quantum applications.

This combination of near-term classical computing benefits and long-term quantum development illustrates how the industry may progress toward practical quantum systems.


Why Quantum Computing Needs Powerful Classical Infrastructure

Quantum computers use quantum bits, or qubits, to represent and manipulate information. Unlike conventional binary bits, which represent either zero or one, qubits can occupy quantum states involving superposition and entanglement.

These properties allow quantum algorithms to approach certain computational problems differently from classical algorithms. Potential applications include molecular simulation, materials research, optimization, and selected problems in cryptography and scientific computing.

However, quantum processors are extremely sensitive systems. Environmental disturbances, imperfect operations, and hardware limitations can introduce errors into calculations. Consequently, a useful large-scale quantum computer must do much more than execute quantum gates. It must continuously monitor and manage the reliability of its operations.


This is where classical computing becomes indispensable.

A fault-tolerant quantum computer requires conventional hardware to process measurement results, identify error patterns, calculate corrections, and coordinate control operations. These activities must happen within tight timing constraints.

In a photonic quantum architecture, information is carried by particles of light, or photons. The system must coordinate optical components, detectors, control electronics, and classical processors. As the number of quantum operations increases, the supporting computational workload can become substantial.

The relationship can be understood through three interconnected layers:

  • Quantum processing: Performs operations using qubits and quantum circuits.

  • Classical control and error correction: Interprets measurement outcomes, detects errors, and coordinates corrections.

  • High-performance computing and software: Provides simulation, algorithm development, workload orchestration, and computational support.

A weakness in any layer can limit the performance of the complete system. Faster quantum hardware alone cannot guarantee useful results if classical processing cannot keep pace.

AMD and PsiQuantum are therefore addressing a system-level engineering problem rather than simply combining processors from two manufacturers.


AMD's Strategic Investment in PsiQuantum

AMD's planned equity investment creates a closer financial relationship between the semiconductor company and PsiQuantum. The companies have not disclosed the investment amount, but the announcement establishes a broader partnership focused on developing infrastructure for utility-scale quantum computing.

The agreement brings together complementary technologies.

AMD develops CPUs, GPUs, adaptive computing devices, and software platforms used across data centers, artificial intelligence, and high-performance computing. PsiQuantum is developing a photonic quantum computing architecture based on silicon photonics.

Under the expanded arrangement, AMD will contribute:

  • EPYC central processing units for general-purpose computing.

  • Instinct graphics processing units for accelerated workloads and large-scale simulation.

  • Versal adaptive system-on-chips for programmable computing and real-time control.

  • ROCm software for GPU programming and workload orchestration.

PsiQuantum will contribute its silicon photonics technology, quantum system architecture, and software and computational workflows.

The investment is strategically significant because it aligns the interests of a major semiconductor supplier with a company attempting to develop a complete quantum computing platform.

For AMD, the partnership offers an opportunity to extend its role into the emerging quantum infrastructure market. For PsiQuantum, access to AMD's computing technologies and engineering capabilities can help address the classical processing requirements of its architecture.

The collaboration also reflects a broader industry trend: quantum computing development increasingly depends on partnerships between specialized quantum hardware companies and established providers of classical computing infrastructure.


PsiQuantum's Silicon Photonics Approach to Quantum Computing

PsiQuantum is pursuing photonic quantum computing, an approach that uses photons as carriers of quantum information.

Photonic systems differ from architectures that use superconducting circuits, trapped ions, or neutral atoms. Instead of relying primarily on electrical or atomic quantum states, photonic quantum computers manipulate light through optical components and integrated photonic circuits.

One attraction of this approach is its potential compatibility with semiconductor manufacturing techniques.

Silicon photonics integrates optical functions with semiconductor-based technology. It allows light to be generated, routed, manipulated, and detected using increasingly sophisticated photonic components.


This creates opportunities to manufacture complex optical systems using techniques related to established chip production.

A photonic quantum computer requires more than an optical processor. It needs a network of components capable of managing photons, routing signals, performing measurements, and coordinating quantum operations.

The architecture must also address practical challenges such as photon loss, detector performance, optical switching, and the reliability of large-scale component integration.

PsiQuantum's approach is built around the ambition of scaling photonic systems toward fault-tolerant quantum computing. Achieving that objective requires coordinating large numbers of optical components with classical computing systems that can respond rapidly to the quantum processor's operations.

AMD's adaptive computing technology and high-performance processors are intended to support this coordination.


Real-Time Quantum Error Correction and the PRISM ML Decoder

Error correction is among the most important requirements for building useful quantum computers.

Quantum information is fragile. Interactions with the surrounding environment and imperfections in hardware operations can introduce errors that compromise calculations.

Classical computers also experience errors, but they can generally use established digital techniques to detect and correct them. Quantum systems face a more complicated problem because directly measuring a qubit can disturb the quantum state being used for computation.

Quantum error-correction techniques address this by encoding logical information across multiple physical qubits. Additional measurements, known as syndrome measurements, provide information about potential errors without directly measuring the encoded logical state.

Classical processors analyze these measurement results to infer what corrections are required.

This creates a demanding computational workflow. Error detection and decoding must occur quickly enough to support the quantum processor's operation. Delayed processing can introduce bottlenecks that limit the speed and scale of the system.

PsiQuantum's PRISM machine-learning decoder is designed to address this challenge.

According to the partnership details, the companies have worked together to adapt PRISM for execution on AMD Versal adaptive system-on-chips.


These devices combine programmable logic with processing capabilities, allowing developers to implement specialized computing functions rather than relying exclusively on general-purpose CPUs.

The objective is low-latency syndrome decoding, enabling classical hardware to process quantum measurement information within the timing requirements of photonic control systems.

The collaboration highlights an important distinction between conventional high-performance computing and real-time quantum control.

Large simulations can benefit from massive parallel processing across GPUs. Error correction, however, can require highly predictable response times and specialized hardware configurations.

The most effective quantum infrastructure will likely combine both approaches, using programmable hardware for time-sensitive control and larger computing systems for simulation, algorithm development, and broader computational workloads.


AMD Instinct GPUs and EPYC CPUs: Building the Quantum-Classical Computing Fabric

The expanded partnership incorporates AMD's broader computing ecosystem.

EPYC CPUs provide general-purpose processing capacity for managing software, coordinating workloads, and handling the conventional computing tasks surrounding quantum operations.

Instinct GPUs provide accelerated computing for highly parallel workloads. GPUs are particularly effective at performing large numbers of mathematical operations simultaneously, making them valuable for scientific simulation, machine learning, and computational chemistry.


The ROCm software stack provides a programming environment for GPU-accelerated applications and helps developers build workloads that can use AMD hardware.

Together, these technologies can support a hybrid computing environment in which classical and quantum processors perform complementary tasks.

Infrastructure component

Technology

Intended role

General-purpose processing

AMD EPYC CPUs

System management and classical computation

Accelerated computing

AMD Instinct GPUs

Simulation, scientific workloads, and computational chemistry

Adaptive hardware

AMD Versal SoCs

Programmable control and real-time error decoding

Software environment

AMD ROCm

GPU programming and workload orchestration

Quantum processing architecture

PsiQuantum silicon photonics

Photonic quantum information processing

Error correction

PRISM ML Decoder

Processing quantum error syndromes

The architecture matters because quantum computers are unlikely to replace classical computing systems.

Instead, they are expected to operate as specialized accelerators within broader computational environments. Classical systems will continue handling tasks for which conventional processors are more practical, while quantum processors may eventually provide advantages for selected computational problems.

The AMD-PsiQuantum partnership is designed around this hybrid model.


Drug Discovery and Computational Chemistry: The Immediate Application

While utility-scale quantum computing remains a longer-term engineering objective, the partnership also targets an application with immediate commercial relevance: pharmaceutical research.

Drug discovery requires scientists to understand how molecules interact, how chemical structures influence biological activity, and how candidate compounds behave under different conditions.

Computational chemistry helps researchers investigate these questions before committing to expensive and time-consuming laboratory experiments.

One area of interest involves covalent inhibitors, a class of molecules relevant to oncology drug discovery. These compounds can form chemical bonds with specific biological targets, potentially influencing the activity of proteins associated with disease.

Predicting molecular interactions accurately is important because small changes in molecular structure can influence binding behavior, selectivity, and potential biological effects.


The companies say PsiQuantum's drug research technology is already running on AMD hardware and can produce more accurate results than existing methods.

They also report initial breakthrough results in the simulation of covalent inhibitors through the combination of PsiQuantum's workflow and AMD's GPU expertise.

This is an important distinction in the partnership's commercial strategy. The near-term value does not depend entirely on the arrival of a fault-tolerant quantum computer.

High-performance classical computing can already support sophisticated molecular modeling and scientific simulation. These capabilities can improve existing research workflows while providing a foundation for future hybrid quantum-classical applications.

The broader opportunity is to develop computational tools that combine several approaches:

  1. Classical simulation for established molecular modeling tasks.

  2. GPU acceleration for computationally intensive calculations.

  3. Machine learning for analyzing complex scientific data.

  4. Quantum algorithms for selected problems when suitable hardware becomes available.

Such a strategy could help pharmaceutical researchers investigate molecular systems that are difficult to model using conventional methods alone.

However, improved computational predictions must ultimately be validated through experimental research. Faster simulations do not independently establish clinical effectiveness or guarantee successful drug development.


Quantum Computing and the Future of Oncology Research

The pharmaceutical industry is a particularly relevant potential market for quantum computing because molecular systems involve complex interactions that can be computationally expensive to model.

Classical computers use numerical approximations to simulate quantum mechanical behavior in molecules. These methods are powerful and widely used, but their computational requirements can increase significantly as molecular complexity grows.

Quantum computers could eventually offer new ways to represent and investigate certain molecular systems.

Potential research applications include:

  • Molecular energy calculations.

  • Chemical reaction modeling.

  • Protein and ligand interaction research.

  • Materials discovery for pharmaceutical applications.

  • Optimization of selected experimental and manufacturing processes.

These applications remain dependent on the development of sufficiently capable quantum hardware and algorithms.

The AMD-PsiQuantum collaboration therefore has a two-stage commercial logic. GPU-powered computational chemistry can support near-term research, while the development of quantum infrastructure prepares the platform for more advanced workloads in the future.

For pharmaceutical companies, this approach could provide a way to explore emerging computational methods without waiting for the complete realization of large-scale quantum computing.


Chicago's Emerging Role in Quantum Infrastructure Development

The partnership also includes a potential regional infrastructure dimension.

PsiQuantum is the anchor tenant for the Illinois Quantum and Microelectronics Park in Chicago. The two companies are evaluating a co-located presence at the site as part of their broader expansion plans.

Co-location could support closer collaboration between hardware engineers, quantum researchers, software developers, and infrastructure specialists.

Building a utility-scale quantum system requires expertise across several disciplines. Photonic engineering, semiconductor technology, control systems, high-performance computing, software development, and scientific applications must work together.

A dedicated quantum campus can provide an environment for developing and integrating these technologies.

The companies also planned to present further technical details about their joint roadmap at the SC26 supercomputing conference in Chicago, scheduled for November 15 to 20, 2026.

The proposed collaboration reflects the increasing overlap between quantum computing research and the established high-performance computing industry.


The Challenges of Building Fault-Tolerant Quantum Computers

Despite the technological promise, several obstacles remain before quantum computing can deliver broad commercial value.

Error correction at scale: Fault-tolerant computing requires managing errors across large numbers of physical qubits. The classical infrastructure must process information quickly and reliably.

Photon loss: Photonic quantum systems must minimize losses during photon generation, transmission, manipulation, and detection.

System integration: A complete quantum computer requires coordination between optical components, control electronics, classical processors, and software.

Manufacturing scalability: Semiconductor-compatible manufacturing may support photonic integration, but producing large, reliable systems remains a substantial engineering challenge.

Software development: Quantum hardware must be supported by algorithms, programming tools, error-correction methods, and workflows that allow researchers and businesses to use it effectively.

Commercial economics: Building and operating advanced quantum infrastructure requires significant capital. Commercial applications must generate enough value to justify these investments.

The AMD partnership addresses some of the classical computing and infrastructure requirements, but it does not eliminate the fundamental challenges of quantum hardware development.

The distinction between a promising architecture and a commercially useful fault-tolerant system remains essential when assessing the industry's progress.


What the AMD-PsiQuantum Partnership Means for the Future of Computing

The expanded partnership reflects a shift in the quantum computing industry toward integrated infrastructure.

Earlier stages of development often emphasized the quantum processor itself. As systems become more sophisticated, the surrounding computing environment is receiving greater attention.

This creates opportunities for established semiconductor companies, high-performance computing providers, software developers, and specialized quantum technology firms.

AMD's involvement demonstrates how conventional computing technologies could become essential components of future quantum systems.

The partnership also points toward a more integrated model of scientific computing, in which CPUs, GPUs, adaptive hardware, and quantum processors work together.


For the wider industry, several developments will be important to monitor:

  • Demonstrated improvements in real-time quantum error decoding.

  • Performance and scalability of the PRISM decoder on AMD adaptive hardware.

  • Practical results from hybrid quantum-classical workflows.

  • Independent validation of computational chemistry improvements.

  • Progress toward larger and more reliable photonic quantum systems.

  • Commercial adoption of the combined infrastructure.

  • Concrete deployment plans associated with the Illinois Quantum and Microelectronics Park.

These milestones will provide a clearer picture of whether the collaboration can translate its technical ambitions into scalable computing infrastructure.


The Quantum Computing Race Is Becoming an Infrastructure Race

AMD's planned equity investment in PsiQuantum and their expanded technology partnership highlight an important reality: the future of quantum computing will depend on much more than quantum processors.

Fault-tolerant systems require classical computing infrastructure capable of handling error correction, control operations, simulation, and software orchestration. AMD's CPUs, GPUs, Versal adaptive devices, and ROCm platform provide technologies intended to support these requirements, while PsiQuantum contributes its silicon photonics architecture and quantum computing expertise.

The partnership also creates opportunities in computational chemistry and oncology research, where GPU-powered workflows can deliver practical value before large-scale quantum hardware becomes commercially available.


For Dr. Shahid Masood and the expert team at 1950.ai, this development illustrates how the next generation of computing may emerge from the convergence of artificial intelligence, high-performance computing, semiconductor engineering, and quantum technology.

The central question is no longer simply how to build a more powerful quantum processor. It is how to construct an integrated system in which quantum hardware and classical infrastructure operate together reliably, efficiently, and at scale.

Key Takeaways

  • AMD announced a planned equity investment in PsiQuantum on November 12, 2026, as part of an expanded quantum infrastructure partnership.

  • The investment amount has not been disclosed.

  • AMD will contribute EPYC CPUs, Instinct GPUs, Versal adaptive computing technology, and the ROCm software stack.

  • PsiQuantum will contribute its silicon photonics platform and quantum computing expertise.

  • The companies are adapting the PRISM machine-learning decoder to AMD Versal hardware for low-latency quantum error correction.

  • The partnership combines quantum processing with classical high-performance computing and adaptive control systems.

  • Computational chemistry and oncology drug discovery are important near-term application areas.

  • The companies report initial results involving the simulation of covalent inhibitors.

  • Both organizations are evaluating a co-located presence at the Illinois Quantum and Microelectronics Park in Chicago.

  • The collaboration highlights the importance of classical computing infrastructure in developing fault-tolerant quantum computers.

Future Outlook

The AMD-PsiQuantum partnership represents a broader transition in quantum computing, from isolated processor demonstrations toward integrated systems designed for practical workloads.

The industry's progress will depend on advances in photonic hardware, error correction, classical processing, software integration, and commercial applications. GPU-accelerated scientific computing may provide valuable capabilities during this transition, while future fault-tolerant quantum systems could expand the range of computational problems researchers can investigate.

If these technologies mature together, the convergence of classical and quantum computing could become an important foundation for scientific discovery, pharmaceutical research, advanced materials, and next-generation computing infrastructure.


Further Reading / External References

AMD to invest in PsiQuantum to advance quantum computing infrastructure

PsiQuantum and AMD Expand Partnership with Equity Investment to Co-Design Utility-Scale Quantum Infrastructure

Comments


bottom of page