Iran's hypersonic missiles and the defense gap they exploit

Iran's Fattah missile series represents a genuine, if often overstated, threat that has exposed critical weaknesses in US and Israeli missile defenses during the ongoing 2026 conflict. Both the Fattah-1 and Fattah-2 have now seen combat use — the Fattah-2 debuting on March 1, 2026, one day after Operations Epic Fury and Roaring Lion commenced — and no fully operational Western system exists today that is specifically designed to intercept maneuvering hypersonic glide vehicles during their glide phase. The defense gap is real, though experts describe it as "difficult but tractable." What follows is the most honest assessment publicly available information allows.

---

What Iran actually has versus what it claims

Iran's hypersonic arsenal centers on two confirmed systems and one unverified third variant. The Fattah-1, unveiled in June 2023 and first combat-deployed during the April 2024 strikes on Israel, is a medium-range ballistic missile with a maneuverable reentry vehicle (MaRV). Iran claims a range of 1,400 km and terminal speeds of Mach 13–15. The Fattah-2, displayed in November 2023 and first used in combat on March 1, 2026, upgrades the design with a true hypersonic glide vehicle (HGV) capable of maneuvering in both pitch and yaw throughout flight, not just in the terminal phase. Iran claims similar speed and a range of approximately 1,500 km. A Fattah-3 combining hypersonic speed with cluster submunitions was reported by a single Lebanese outlet on March 14, 2026, but has not been independently verified.

The critical question is whether these are genuinely "hypersonic" in the way military analysts use the term. Fabian Hinz of the International Institute for Strategic Studies (IISS), the most cited Western analyst on this topic, assessed the Fattah-1 as "obscuring more than it illuminates" — essentially an MRBM with an extra solid rocket motor in its reentry vehicle that "can do some basic maneuvers, but not for the same amount of time and not as dramatic" as US, Russian, or Chinese systems. Many conventional ballistic missiles reach Mach 13+ during terminal descent; the differentiator is sustained maneuvering at hypersonic speed, which the Fattah-1 has limited ability to perform. The Fattah-2's HGV design is more credible as a true hypersonic weapon, capable of atmospheric skip-glide flight and approaching targets from unexpected angles, but it was only deployed for the first time weeks ago and independent technical verification remains sparse.

Iran Watch, a Wisconsin Project publication drawing on IISS and Congressional Research Service analysis, placed the Fattah "largely in a class of its own" — neither a classic HGV nor a hypersonic cruise missile, but something more modest that nonetheless complicates defense planning significantly. Iran also fields the Khorramshahr-4 (range ~2,000 km, 1,500 kg warhead) with claimed HGV-like capabilities, though these are similarly debated. Suspected technology transfer from Russia and possibly North Korea — whose Hwasong-16B featured an HGV warhead in 2024 — may explain the rapid pace of Iran's development.

---

Combat use in Operations Epic Fury and Roaring Lion

The conflict that began on February 28, 2026 — when joint US-Israeli strikes killed Supreme Leader Ali Khamenei and struck Iran's military infrastructure — has produced the first large-scale battlefield testing of Iran's hypersonic arsenal. The US designated its campaign Operation Epic Fury; Israel named its parallel operation Operation Roaring Lion. Both are confirmed by CENTCOM, the White House, CSIS, and major international media.

Iran's response has been massive. By March 15, Iran claimed it had fired approximately 700 missiles and 3,600 drones at Israeli, US, and allied targets across the region, including bases in Bahrain, Jordan, Kuwait, Qatar, Saudi Arabia, Turkey, and the UAE. Within this barrage, Iran deployed Fattah-1 missiles beginning February 28 and Fattah-2 missiles beginning March 1 — the Fattah-2's combat debut. Military Watch Magazine reported footage showing at least three successful Fattah-2 impacts, including one on a fortified IDF command center that reportedly killed seven senior officers. The IRGC claimed the Fattah-2 evaded multiple interceptors during strikes near Tel Aviv, the Israeli Ministry of Defence, and Ben Gurion Airport.

These claims require significant caution. Defense analysts note that isolated missile impacts do not necessarily demonstrate defense system failure — defenders routinely prioritize protecting populated areas over hardened military facilities when interceptor stocks are limited. The US Department of Defense has not confirmed hypersonic missile use in reported strikes. An IRGC spokesperson stated on March 16 that "most of the IRGC's weapons cache remains intact" and that missiles used so far are from "a decade ago," with newer missiles not yet fired — a claim that may be strategic messaging rather than literal truth.

This conflict follows two precedent-setting Iranian attacks in 2024. In April 2024 (Operation True Promise I), a coalition defense effort achieved approximately 99% interception of ~320 incoming weapons, though 6–10 ballistic missiles struck Nevatim Airbase. In October 2024 (True Promise II), ~200 ballistic missiles including confirmed Fattah-1s achieved a lower interception rate of roughly 75%, with over two dozen missiles penetrating defenses. The June 2025 Twelve-Day War saw Israel claim an 86% interception rate against Iranian ballistic missiles — respectable, but meaning one in seven missiles reached their targets.

---

The defense landscape has critical holes

No existing missile defense system was purpose-built to intercept a maneuvering hypersonic glide vehicle during its atmospheric glide phase. Current defenses can be grouped by their actual (rather than theoretical) capability against hypersonic threats.

The SM-6 Sea Based Terminal Increment 3, certified in August 2025 for deployment on Aegis destroyers, is what former MDA Director VADM Jon Hill called "really the nation's only hypersonic defense capability" — though he emphasized its capability remains "nascent." It uses a blast-fragmentation warhead for terminal-phase intercept and demonstrated simulated engagement of a hypersonic target vehicle in March 2025, but has not conducted a live intercept of a hypersonic weapon. The Patriot PAC-3 MSE achieved the first verified "hypersonic" intercept in May 2023, downing a Russian Kinzhal over Kyiv, but the Kinzhal behaves more like a maneuverable ballistic missile than a true HGV with sustained glide-phase maneuvering.

THAAD, deployed to Israel after the October 2024 attacks, operates at 40–150 km altitude — above where HGVs fly during their glide phase (20–70 km). European Security & Defence assessed it as "ill-suited for HCM/HGV interceptions" in its current configuration. A critical radar upgrade — the AN/TPY-2 with gallium nitride antenna, delivered May 2025 — doubles detection range and can track separated HGVs, but the interceptor itself remains altitude-mismatched. THAAD 6.0, with new interceptor capabilities, is not expected until 2027. Israel's Arrow 3 operates exoatmospherically and can potentially engage ballistic missiles before HGV separation, but cannot follow an HGV into its atmospheric glide.

The two systems specifically designed for hypersonic defense remain years away. The US Glide Phase Interceptor (GPI), a joint program with Japan featuring a re-ignitable motor and multimode seeker, faces a roughly three-year delay due to funding cuts — delivery unlikely before ~2035. Israel's Arrow 4, developed jointly with MDA and featuring AI-enhanced guidance for countering maneuvering HGVs, is further along, with live trials beginning mid-2026 and possible early deployment that year. It represents the most advanced Western hypersonic interceptor approaching operational status. Rafael's independently funded SkySonic interceptor targets the same threat set but has not received Israeli MoD funding.

System	Anti-hypersonic capability	Status	Key limitation
SM-6 SBT Inc. 3	Moderate (terminal phase)	Deployed Aug 2025	No live hypersonic intercept yet
Patriot PAC-3 MSE	Limited (combat-proven vs. Kinzhal)	Operational	Short engagement window against true HGVs
Arrow 4	High (purpose-built)	Live trials mid-2026	Not yet fielded
GPI	High (purpose-built)	Development	~2035 delivery, severely delayed
THAAD	Very limited	Operational	Altitude mismatch with glide phase
Arrow 3	Limited (exoatmospheric only)	Operational	Cannot engage in atmospheric glide

---

Analysts agree the gap is real but not permanent

The most comprehensive expert assessment comes from CSIS Missile Defense Project directors Tom Karako and Masao Dahlgren, who concluded that "defending against hypersonic missiles is strategically necessary, technologically possible, and fiscally affordable, but it will not be easy." They emphasized that the "current Ballistic Missile Defense System, largely equipped to contend with legacy ballistic missile threats, must be adapted to this challenge." James Acton of the Carnegie Endowment noted that point-defense systems "could very plausibly be adapted" but "can only defend small areas" — defending large territories would require unaffordable numbers of batteries.

The detection gap may be the most acute vulnerability. Former Under Secretary of Defense Mike Griffin warned that hypersonic targets are "10 to 20 times dimmer" than what US geostationary satellites normally track, while terrestrial radars face line-of-sight limitations against low-flying glide vehicles. The HBTSS space sensor demonstrated tracking of a maneuvering hypersonic target in March 2025, but the full Proliferated Warfighter Space Architecture remains years from completion. The command-and-control gap compounds this: CRS has assessed that the current architecture "would be incapable of processing data quickly enough to respond to and neutralize an incoming hypersonic threat."

Perhaps most critically for the ongoing conflict, the inventory gap looms large. The Soufan Center assessed on March 1, 2026 that "a war of attrition that exhausts missile defense inventories is the most beneficial outcome for Tehran." Each Arrow 3 or SM-3 interceptor costs tens of millions of dollars; Iran's strategy of firing cheaper conventional missiles in volume alongside select hypersonic weapons forces defenders into unfavorable cost-exchange ratios. The April 2024 defense of Israel cost approximately $1 billion against an Iranian attack costing $80–100 million.

---

Strategic implications extend well beyond the battlefield

Iran's hypersonic capability — even in its technically modest current form — functions as what TRENDS Research & Advisory analyst Dr. Jean-Loup Samaan calls "a threat multiplier rather than a game changer." The strategic implications operate on multiple levels.

The most immediate is compressed decision timelines. Iranian MRBMs reach Israel in 6–10 minutes; a maneuvering hypersonic weapon within that envelope leaves defenders almost no margin for error. Combined with Iran's demonstrated tactic of integrating drones, cruise missiles, and ballistic missiles in simultaneous time-on-target attacks — what MDA Director Collins called "larger than we've seen ever" after April 2024 — hypersonic weapons force defenders to allocate scarce interceptors under extreme time pressure against an unpredictable target. Rafael's VP Yuval Baseski likened defending against a hypersonic missile to "defending LeBron James with a single player."

Iran's approach reflects deliberate asymmetric strategy. The National Security Journal assessed that hypersonic missiles "fit neatly into Iran's asymmetric warfare approach, offering a means to strike quickly and decisively while avoiding interception." Tehran gains deterrent value from ambiguity — Army Recognition noted in March 2026 that "Iran is deliberately blurring missile identities to complicate attribution and amplify perceived deterrent value." The psychological impact of weapons perceived as unstoppable carries strategic weight independent of actual interception rates.

An emerging regional arms race adds longer-term concern. Saudi Arabia is reportedly intensifying efforts to acquire hypersonic technology, potentially through Russian partnerships. Turkey is building indigenous ballistic and hypersonic capabilities. Israel's massive post-2025 investment in Arrow production (tripled rate), Arrow 4 development, Iron Beam lasers, and David's Sling upgrades represents its own accelerated response. Reports — sourced to ISPI but requiring further verification — that Khamenei authorized miniaturized nuclear warhead development in October 2025 would, if true, dramatically alter the strategic calculus by marrying hypersonic delivery systems to nuclear payloads.

Technology transfer remains a serious concern. Following the June 2025 war, Iran accelerated negotiations with Russia and China for advanced systems. Defence Security Asia reported discussions about Chinese HGV technology transfer, while China supplied over 2,000 tons of sodium perchlorate for solid propellant production despite reinstated UN sanctions. The speed of Iran's hypersonic development has led multiple analysts to conclude it likely received significant external assistance.

---

Conclusion

The honest assessment is one of uncomfortable ambiguity. Iran's "hypersonic" weapons are less sophisticated than Tehran claims — the Fattah-1 is essentially a ballistic missile with enhanced terminal maneuvering, not a peer to Russia's Avangard or China's DF-ZF. But this distinction matters less than it might seem. Even technically modest maneuvering capability at hypersonic speeds, combined with Iran's proven tactic of saturating defenses with mixed-threat salvos, exploits a genuine gap in Western defenses for which no purpose-built operational interceptor yet exists. The SM-6 offers nascent capability; Arrow 4 and GPI represent the real solutions, but neither will be fully operational before 2027 at the earliest.

The ongoing conflict is generating unprecedented real-world data on hypersonic attack and defense, but that data remains contested and difficult to verify amid active hostilities. What is clear is that the strategic landscape has shifted: the era in which missile defense could provide near-total protection — as in the 99% interception of April 2024 — has given way to one where defenders must accept some degree of leakage and prioritize accordingly. The critical unknown is whether Iran's stockpile of advanced missiles is large enough, and its production capacity resilient enough after successive rounds of strikes, to sustain the kind of attrition strategy that would genuinely overwhelm allied defenses. That question will likely be answered in the weeks ahead.

Message from Charles Aulds

"Does Donald Trump expect a handful of European frigates to do what the powerful US Navy cannot? This is not our war, and we did not start it!" 

___

German Defense Minister Boris Pistorius

"While taking the necessary action to defend ourselves and our allies, we will not be drawn into the wider war. We will keep working towards a swift resolution that brings security and stability back to the region and stop the Iranian threat to its neighbours."

___

Keir Starmer, Prime Minister of the United Kingdom

Donald Trump has backed himself into a corner in the Strait of Hormuz, and he knows it. 

___

The Daily Adda

 

Donald Trump asked the world to join a multinational naval coalition to keep the Strait of Hormuz open. 

The results, as of this morning, when France gave it's reply:

🇫🇷 France: Rejected

🇮🇹 Italy: Rejected

🇪🇸 Spain: Rejected

🇯🇵 Japan: Rejected

🇳🇴 Norway: Rejected

🇨🇦 Canada: Rejected

🇦🇺 Australia: Rejected

🇩🇪 Germany: Rejected

🇬🇧 UK: Rejected

🇨🇳 China: No response 

🇳🇱 Netherlands: No response

🇰🇷 South Korea: No response

Photonic Neuromorphic Learning Networks: State of the Field (2025–2026)

Executive Summary

Photonic neuromorphic learning networks — computing systems that use light rather than electrons to implement brain-inspired neural architectures — have crossed a series of landmark thresholds between 2024 and early 2026. The field has moved from single-layer proof-of-concept demonstrations to multi-layer, on-chip systems capable of in-situ gradient-descent training, all-optical nonlinear activation, and real-time reinforcement learning. Simultaneously, massive commercial interest — including NVIDIA's $4 billion investment in photonics firms in March 2026 — signals that the technology is transitioning from academic labs toward infrastructure-scale deployment.

Why Photonics for Neuromorphic Computing?

Traditional von Neumann electronic architectures face fundamental barriers in AI workloads: memory-compute separation creates energy-expensive data movement, transistor scaling is approaching physical limits, and sequential processing constrains parallelism. Photonic systems offer a set of intrinsic advantages that are uniquely suited to neural network computation:[1][2]

• Ultrafast operation — optical signals propagate at the speed of light, enabling sub-nanosecond inference latencies
• Ultra-low energy propagation — photons carry information with minimal resistive loss
• Massive parallelism — wavelength-division multiplexing (WDM) allows multiple data streams to coexist on the same waveguide simultaneously
• High bandwidth — optical channels support data rates far exceeding electronic interconnects
• Low electromagnetic interference — photonic signals are immune to the crosstalk that plagues dense electronic circuits

The neuromorphic layer adds brain-inspired architectural principles: sparse event-driven computation, asynchronous processing, and local synaptic plasticity rules. Together, these properties make photonic neuromorphic systems candidates for both edge AI (autonomous vehicles, robotics) and data-center-scale inference acceleration.[3][4]

Key Architectural Paradigms

Mach-Zehnder Interferometer (MZI) Mesh Networks

The dominant architecture for integrated photonic neural networks (PNNs) uses programmable MZI meshes to implement matrix-vector multiplication in the optical domain. Each MZI functions as a tunable beamsplitter (a two-neuron synapse), and cascaded meshes of N×N MZIs can represent arbitrary unitary linear transformations. Mach-Zehnder meshes have formed the backbone of virtually every on-chip PNN demonstration from the foundational work onward, and they remain central to the latest 2025–2026 results.[5][6]

The current frontier is achieving end-to-end integration: pairing the MZI weight matrix with on-chip nonlinear activation and on-chip gradient computation so that training itself can occur within the optical domain, rather than offloading gradients to a host CPU.

Spiking Photonic Neural Networks (S-PNNs)

Spiking neural networks encode information in the timing and rate of brief pulses rather than continuous amplitudes, closely mimicking biological neurons. In the photonic domain, spiking behavior is generated using semiconductor lasers with saturable absorbers, which exhibit threshold-triggered optical pulse emission analogous to neuronal action potentials.[7][6]

A landmark result published in Optica in March 2026 by researchers at Xidian University demonstrated the first large-scale programmable incoherent photonic spiking neural network (PSNN) capable of performing both linear and nonlinear computation entirely in the optical domain — eliminating the electronic back-conversion step that previously nullified photonic speed advantages. The two-chip system comprised:[8][6]

• A 16×16 Mach-Zehnder interferometer mesh chip with 272 trainable synaptic parameters, enabling simultaneous processing of 16 optical channels
• A distributed feedback (DFB) laser array chip with saturable absorbers, optimized for low-threshold nonlinear spiking activation

The system used a hardware-software collaborative training framework: initial global training in software, followed by on-chip fine-tuning to compensate for fabrication-induced device variations. The deployed system successfully mastered two classic reinforcement learning benchmarks — CartPole (balancing a pole on a cart) and Pendulum (swinging an inverted pendulum to vertical) — through trial-and-error learning in real time. The team identified a 128-channel fully integrated chip as the next target, sufficient for neuromorphic autonomous navigation tasks.[9][10][8]

Diffractive Optical Neural Networks (D²NNs)

An orthogonal approach encodes the neural network weights into the diffraction properties of cascaded transmissive layers rather than on-chip waveguides. Incident light propagates through optimized phase masks, performing the equivalent of hidden-layer processing as it diffracts between planes. This architecture is inherently passive (no power required for inference) and can in principle operate at the speed of light with zero dynamic energy consumption during inference.[11]

Recent work published in Science Advances (2025) introduced directional-diffractive deep neural networks (D-D²NNs), which encode the wave propagation direction as an additional degree of freedom, enabling simultaneous multi-channel classification and high-capacity data encryption using metasurfaces. A separate study (February 2026) demonstrated anti-interference D²NNs that achieved 87.4% accuracy on multi-object recognition tasks under 40 categories of dynamic interference — addressing a key weakness of earlier D²NN systems that struggled with real-world occlusion and clutter.[12][13]

Reservoir Computing

Photonic reservoir computing uses a fixed, high-dimensional dynamical system (the "reservoir") as a nonlinear projection layer, with only the output weights trained. This dramatically reduces training complexity while preserving the temporal dynamics needed for sequence processing. March 2025 work published in Optica demonstrated graded photonic neurons using quantum-dot (QD) gain sections — overcoming the binary-only limitation of earlier photonic spiking neurons — and integrated multiple QD lasers on a single die for reservoir use. The system achieved 98% accuracy on arrhythmia detection and 92% on handwriting classification, with 100 GHz QD lasers enabling sub-nanosecond temporal resolution.[7]

A 2025 study in Optics and Laser Technology extended optoelectronic reservoir computing with enhanced readout mechanisms, demonstrating improved performance on benchmark time-series prediction tasks. A November 2025 arXiv preprint explored spin-orbit coupling in organic crystal resonators as a photonic reservoir medium, using a 2D hexagonal photonic crystal structure for low-cost, easily fabricated neuromorphic computation.[14][15]

The Training Problem: In-Situ Learning Breakthroughs

The central challenge for photonic neural networks has historically been training: how to compute weight gradients when the forward pass is optical. Early demonstrations trained weights digitally and then programmed the results onto the photonic chip (offline training), which is adequate for fixed inference but precludes online learning or adaptation to hardware imperfections.

In-Situ Backpropagation

The theoretical framework for optical backpropagation was established by Hughes et al. (2018) using adjoint variable methods, showing that gradients could be obtained by measuring optical intensity within the device. Experimental realization came in a landmark 2023 Science paper by Pai et al. at Stanford, who demonstrated in-situ backpropagation on a three-layer, four-port silicon PNN using programmable phase shifters and optical power monitors. By interfering forward- and backward-propagating light, the chip computed backpropagated gradients without any electronic gradient computation, achieving >94% test accuracy on classification tasks, comparable to digital simulations.[16][17][18]

End-to-End On-Chip Gradient Descent

Building on the in-situ backpropagation foundation, a June 2025 preprint from Nokia Bell Labs (Ashtiani et al.) demonstrated the first integrated photonic deep neural network with end-to-end on-chip gradient-descent backpropagation training. All linear and nonlinear computations occur on a single chip, producing training that is robust to fabrication-induced device variations — historically a major obstacle to scaling. The demonstrated accuracy matched ideal digital model performance on two nonlinear classification tasks.[19]

A September 2025 paper in Light: Science & Applications (Wu et al.) demonstrated end-to-end two-class classification of fashion images and four-class classification of handwritten digits using a purely on-chip optical architecture, overcoming the cascadability and coherence-source limitations that had previously constrained network depth and input size.[20][5]

The Nonlinearity Problem: Solved with Programmable Activation

A persistent barrier to deep photonic networks has been the absence of reconfigurable, strong optical nonlinearity. Without nonlinear activation functions, photonic networks are limited to solving linearly separable problems and cannot compete with multi-layer electronic networks on real-world tasks.

In April 2025, University of Pennsylvania engineers reported the first field-programmable photonic chip capable of delivering tunable nonlinear activation in the optical domain, published in Nature Photonics. The architecture uses InGaAsP semiconductor material: a pump beam controls the spatial distribution of excited charge carriers, reshaping the signal beam into high-order polynomial functions. Key results included:[21][22][23]

• Polynomial nonlinear networks outperformed equivalent linear photonic networks on iris species classification (96% vs. 86% accuracy)
• The platform is fully reconfigurable — different nonlinear functions can be programmed without hardware modification
• Team leader Liang Feng described it as "a true proof-of-concept for a field-programmable photonic computer"

This result is architecturally significant because it means photonic chips can now be reprogrammed not just in their linear weight matrices (as MZI meshes permit) but also in the shape of their activation functions — approaching the flexibility of software-defined electronic networks.

Memory Integration: Eliminating the von Neumann Bottleneck

A February 2026 paper in Nature Communications (Lam, Khaled, Bilodeau et al., from UBC, Queen's University, and Princeton) addressed the energy cost of data movement between memory and photonic compute units — the photonic analog of the von Neumann bottleneck. The solution: co-locate analog electronic memory directly with photonic computing units on the same monolithic chip.[24][25]

The key results of this electro-optic analog memory architecture:[25][26]

• >26× power savings compared to conventional SRAM-DAC architectures
• In-situ training and inference validated on MNIST with >90% accuracy
• Leaky analog memories (non-ideal retention) remain viable provided the retention-to-latency ratio exceeds 100, enabling more relaxed fabrication tolerances
• Eliminates the energy cost of repeated analog-to-digital and digital-to-analog conversion

This memory co-location strategy is a direct parallel to in-memory computing approaches in electronic neuromorphic chips (e.g., IBM's NorthPole), applied to the photonic domain.

Phase-Change Material (PCM) Non-Volatile Weights

For applications requiring persistent weight storage without any static power consumption, phase-change materials (PCMs) — particularly Ge₂Sb₂Te₅ (GST) and the lower-loss Ge₂Sb₂Se₅ (GSSe) — offer non-volatile optical memory by switching between amorphous and crystalline phases with distinct refractive indices. A 2023 study demonstrated a GSSe-based photonic RAM achieving 0.12 dB total insertion loss for a 4-bit memory with 500,000 write-erase cycles, representing a 100× improvement in signal-to-loss ratio over GST devices.[27][28]

PCM-based weights are particularly attractive for neuromorphic applications because they consume zero static power during inference (the phase state persists indefinitely without refresh). However, fabrication process variations introduce phase noise that currently limits scalable silicon photonic neural networks with PCM phase shifters to approximately N=16 before accuracy degrades significantly — a fabrication challenge requiring pre- or post-fabrication correction methods.[29]

Scaling and Architecture Innovations

Incoherent Architectures

Most early PNN demonstrations used coherent light sources, which require strict phase control and temperature stabilization across the chip — challenging to maintain at scale. The March 2026 Xidian University result specifically demonstrated that incoherent photonic neuromorphic computation is viable for spiking networks, relaxing the coherence requirements and significantly improving practical scalability.[6][10]

3D Stacking

A 2025 paper extended deep neural network training to a 3D-stacked photonic architecture (LSPA), using multi-layered non-volatile photonic-PCM cells in a high-density computational fabric. The system was evaluated on VGG-16, ResNet-50, GoogLeNet, Transformer, GNMT, and LLaMA 7B — representing a significant expansion in the scale of workloads addressable by photonic architectures.[30]

PDONN: Overcoming Depth Limits

A October 2025 result (EurekaAlert) introduced the Progressive Diffractive Optical Neural Network (PDONN) architecture, which uses architectural innovation to overcome the optical nonlinear activation cascadability problem that previously limited end-to-end on-chip ONN depth. The approach uses more accessible, non-coherent light sources while maintaining competitive accuracy on large-scale AI tasks, removing two of the three principal obstacles (depth, coherence, input scale) to practical integration.[20]

Applications and Demonstrations

Application	Architecture	Key Result	Source
Reinforcement learning (CartPole, Pendulum)	Spiking PNN, 16-ch MZI + DFB	Real-time all-optical RL learning	[8][6]
Image classification (MNIST, Fashion)	End-to-end on-chip ONN	Accuracy matches digital model	[19][5]
Arrhythmia detection	QD reservoir computing	98% accuracy at 100 GHz	[7]
Iris species classification	Field-programmable nonlinear PNN	96% accuracy vs. 86% linear	[21]
Multi-object recognition (THz)	Anti-interference D²NN	87.4% under 40 interference categories	[13]
AI inference acceleration	Silicon PICs (HP Labs)	Wafer-scale III-V integration	[4]

Commercial Ecosystem and Investment

The research momentum has attracted substantial commercial investment. On March 2, 2026, NVIDIA announced $2 billion investments in each of Lumentum and Coherent — totaling $4 billion — alongside multibillion-dollar purchase commitments to advance silicon photonics and optical interconnect manufacturing for AI data centers. Jensen Huang framed the move as establishing "next generation gigawatt-scale AI" infrastructure.[31][32][33][34]

While these investments are primarily directed at optical interconnects (data movement between chips) rather than neuromorphic computing per se, they accelerate the maturation of the photonic manufacturing supply chain — lithography, laser integration, packaging — on which neuromorphic PNNs directly depend.[35][34]

At the national policy level, the Netherlands launched a 10–30 year neuromorphic computing roadmap in October 2025, coordinating universities, industry, and government around a vision that includes photonic neuromorphic systems as a strategic pillar alongside quantum technologies and edge AI.[36][37]

Market analysts estimate the neuromorphic computing market at approximately USD 28.5 million in 2024, projected to reach USD 1.32 billion by 2030 at a CAGR of 89.7%, driven primarily by edge computing demands in autonomous driving, industrial automation, and real-time sensor fusion.[3]

Open Challenges

Despite the rapid progress, several engineering and theoretical challenges remain before photonic neuromorphic learning networks achieve broad deployment:

• Fabrication tolerances: Waveguide and phase-shifter variations accumulate across chip area, degrading accuracy. Current MZI-based networks require post-fabrication calibration; PCM-based networks show accuracy collapse beyond N≈16 without correction.[29]
• Nonlinear activation at scale: While the Penn InGaAsP chip demonstrates programmable nonlinearity, integrating high-order polynomial activation with large MZI weight meshes in a single monolithic process remains unsolved.
• Optical fan-in/fan-out: Connecting neurons with all-to-all or structured sparsity patterns in integrated photonics is far more constrained than in electronics; review papers identify optical fan-in and fan-out as a principal remaining bottleneck.[2]
• Scalability of spiking systems: The March 2026 demonstration used a 16-channel chip; the team identified 128 channels as necessary for real navigation tasks, with full hybrid integration still unrealized.[8]
• Training algorithms: In-situ backpropagation has been experimentally demonstrated on small networks (3–4 layers, <100 ports); scaling gradient propagation to hundreds of layers remains an open problem. Direct feedback alignment (DFA) is under investigation as a more parallelizable photonic training alternative.[38]
• Analog memory drift: Even with electro-optic analog memory co-location, leaky memory retention sets a constraint on how slowly learning can proceed relative to network latency.[25]

Outlook

The period from late 2024 through early 2026 represents a phase transition in photonic neuromorphic learning — from demonstrating that the physics is possible to demonstrating that it is practically trainable, scalable, and applicable to real tasks. Three threads are converging: (1) on-chip gradient computation making self-training chips a reality; (2) programmable nonlinearity breaking the depth barrier; and (3) analog memory co-location eliminating the dominant energy cost. The next 2–3 years are likely to see 128-channel spiking chips, fully hybrid-integrated edge modules, and first commercial neuromorphic photonic accelerators for inference at the edge — driven by both the maturing research base and NVIDIA-scale industrial investment in the underlying photonic manufacturing ecosystem.[34][35][8]

References

• Neuromorphic Photonics - These novel Neuromorphic Photonic systems for information processing will be able to provide very hi...
• Integrated platforms and techniques for photonic neural ... - In this review, we present a comprehensive introduction to the three fundamental building blocks of ...
• Growth Opportunities in Neuromorphic Computing 2025- ... - The neuromorphic computing market was worth approximately USD 28.5 million in 2024 and is estimated ...
• 2025 IEEE Study Leverages Silicon Photonics for Scalable ... - This unique photonic platform can achieve wafer-scale integration of all of the various devices requ...
• Scaling up for end-to-end on-chip photonic neural network ... - by B Wu · 2025 · Cited by 14 — We successfully demonstrate end-to-end two-class classification of fa...
• Photonic chips advance real-time learning in spiking ... - 05 March 2026. Photonic chips advance real-time learning in spiking neural systems. New approach ena...
• High-Speed Reservoir Computing With Integrated Laser ... - So-called neuromorphic computing involves the use of physical artificial neurons to do computing in ...
• Photonic chips advance real-time learning in spiking ... - New approach enables fast, all-optical learning and decision making, paving the way for robots that ...
• Photonic Chips Propel Real-Time Learning in Spiking ... - By demonstrating the feasibility of large-scale photonic spiking neurons with hardware-in-the-loop t...
• Photonic Chips Boost Real-Time Learning in Neural Nets - This breakthrough in photonic computing chips could lead to significant improvements in the speed, e...
• Computer Imaging Based on Optical Diffractive Neural ... - Diffractive deep neural networks (D2NNs) represent the cutting edge of this renaissance, utilizing l...
• High-capacity directional information processor using all ... - by G Wang · 2025 · Cited by 2 — We propose a directional-diffractive deep neural network (D-D2NN) by...
• Anti-interference diffractive deep neural networks for multi ... - By integrating multi-dimensional optical encoding technology with AI D2NN, each channel can be dedic...
• Reservoir neuromorphic computing based on spin-orbit ... - In this work, we present a simple, cheap and easy to fabricate implementation of a physical reservoi...
• Advancing optoelectronic reservoir computing: enhancing ... - by LMS Dias · 2025 · Cited by 1 — Reservoir computing is a neuromorphic architecture based on artifi...
• Experimentally realized in situ backpropagation for deep learning ... - by S Pai · 2023 · Cited by 353 — We experimentally trained a three-layer, four-port silicon photonic...
• Backpropagation training achieved in photonic neural network - Neural networks made from photonic chips can be trained using on-chip backpropagation – the most wid...
• Training of photonic neural networks through in situ backpropagation - by TW Hughes · 2018 · Cited by 645 — In this work, we introduce a method that enables highly efficie...
• Integrated photonic deep neural network with end-to- ... - Here, we report the demonstration of an integrated photonic deep neural network with end-to-end on-c...
• Researchers overcome key scaling barriers in photonic AI ... - Photonic neural networks (ONNs) are considered a competitive alternative due to their significant ad...
• Photonic Chips Go Nonlinear - Researchers at the University of Pennsylvania, USA, have developed what they believe to be the first...
• Programmable, All-Optical, Photonic Architecture for ... - This approach allows the implementation of polynomial nonlinear functions of various orders, enablin...
• Programmable photonic chip uses light to accelerate AI ... - Penn Engineers have developed the first programmable chip that can train nonlinear neural networks u...
• Neuromorphic photonic computing with an electro-optic ... - We propose an analog electronic memory co-located with photonic computing units to eliminate repeate...
• Neuromorphic Photonic Computing with an Electro-Optic ... - We propose an analog electronic memory co-located with photonic computing units to eliminate repeate...
• Neuromorphic Photonic Computing with an Electro-Optic ... - Our analysis shows that integrating analog memory into a neuromorphic photonic architecture can achi...
• Electrical programmable multilevel nonvolatile photonic ... - by J Meng · 2023 · Cited by 103 — Here we demonstrate a multistate electrically programmed low-loss ...
• Neuromorphic Photonics Based on Phase Change Materials - by T Li · 2023 · Cited by 30 — This work reviews the most promising neuromorphic devices based on PC...
• PCM-based Silicon Photonic Neural Networks under ... - by A Shafiee · 2025 · Cited by 1 — ABSTRACT. In this paper, we analyze optical phase noise due to fa...
• Extending Energy-Efficient and Scalable DNN Training and ... - LSPA employs multi-layered 3D stacking of non-volatile photonic-PCM cells, creating a high-density c...
• Nvidia to invest $4B into photonics companies Coherent, Lumentum - Coherent and Lumentum will both receive $2 billion each from the chip giant as part of the strategic...
• Nvidia to invest $2 billion each in Lumentum, Coherent to bolster AI ... - Nvidia will invest $2 billion each in photonic product makers Lumentum and Coherent , as it looks to...
• Nvidia Invests US$4 Billion in Photonic Technology - Lumentum and Coherent will each receive US$2 billion to support AI infrastructure development.
• NVIDIA's $4B Optics Bet Signals Photonics as AI's Next Bottleneck - NVIDIA invests $4B in optical interconnect supply chain with Coherent and Lumentum to address silico...
• Nvidia's Big Investment In Photonics While Prepping Vera Rubin Chips - On the photonics side, Techradar reports Lumentum stock rose 5% and Coherent value spiked 9% on the ...
• Neuromorphic Computing Roadmap 2025 - Technological potential: Neuromorphic computing can strengthen the position of leading technologies ...
• The Netherlands sets course: a Roadmap to strengthen its ... - The Netherlands has officially launched its national roadmap for neuromorphic computing, commissione...
• In situ training with silicon photonics neural networks - This talk will summarize our recently proposed silicon photonic architecture [21] that uses an elect...