A faster, more accurate way to explore innovation data—now available in Cypris.
For innovation teams, speed and accuracy aren’t optional—they’re critical. You need to quickly find all relevant documents, slice and dice datasets however you want, and trust that the results are complete and representative. With this in mind, we’ve upgraded how semantic search works inside Cypris.
Today, we’re launching an upgraded search infrastructure that gives users access to full, exact result sets—unlocking more powerful analysis, faster iteration, and deterministic filtering and charting.
Unlike traditional semantic or vector search engines—which make it difficult to count, filter, or chart large sets of matched documents—our new approach prioritizes transparency and performance while preserving semantic relevance.
Why we moved away from vector search
Our original implementation relied on semantic and vector search to capture the “meaning” behind user queries. But as our platform evolved, it became clear that these systems weren’t well-suited for our core use cases.
Users needed:
- Deterministic filtering (e.g., "how many results match this atom?")
- Transparent, complete result sets to power charts and dashboards
- Fast, repeatable queries that don’t change subtly over time
Modern vector search systems don’t easily support this level of transparency. They return approximate matches and abstract similarity scores, often making it hard to understand why a document was returned—or whether it’s the full picture.
So we made a decision: move away from vector search and lean into what traditional search engines do best.
A return to boolean and lexical search—with a twist
We rebuilt our search infrastructure on top of Elasticsearch’s powerful boolean and lexical search capabilities. This shift brings major advantages:
- Faster query speeds that dramatically improve iteration time
- Deterministic filtering and counts, so every chart is grounded in the full dataset
- Predictable, explainable results that users can trust
But we didn’t stop there.
To preserve the benefits of semantic understanding, we’ve rethought where that intelligence should live—not at query time, but at data ingestion.
Capturing semantic meaning at ingest time
Instead of computing document-query similarity during search, we enrich documents at the time of ingestion. Here’s how:
- Synonym expansion: We find related words and concepts not explicitly mentioned in the document and add them as fields, enabling semantic-style recall via lexical search.
- Stemming: Both queries and documents are reduced to their root forms, allowing consistent matches (e.g., “running” and “run”).
The result? You get the same functionality—semantically relevant results—without the opacity or latency tradeoffs of vector search.
What’s next: Reranking for even better relevance
We’re not done. Coming soon to Cypris is a reranking layer that boosts the most relevant results to the top of the list using lightweight vector techniques.
Here’s how it works:
- A standard lexical search retrieves the full result set.
- We take the top N results and rerank them using vector similarity, powered by Elasticsearch’s new hybrid scoring capabilities.
- You get faster queries with even better relevance—without compromising on counts or transparency.
This layered approach gives us the best of both worlds: precise filtering and fast queries, plus smarter ordering of results where it matters most.
We’re excited to bring this upgrade to our users, and we’re already seeing teams iterate faster and uncover insights more confidently. This is a foundational shift—and just the beginning of what’s to come.
Want a walkthrough of what’s changed? Reach out to our team.
Introducing our upgraded semantic search
A faster, more accurate way to explore innovation data—now available in Cypris.
For innovation teams, speed and accuracy aren’t optional—they’re critical. You need to quickly find all relevant documents, slice and dice datasets however you want, and trust that the results are complete and representative. With this in mind, we’ve upgraded how semantic search works inside Cypris.
Today, we’re launching an upgraded search infrastructure that gives users access to full, exact result sets—unlocking more powerful analysis, faster iteration, and deterministic filtering and charting.
Unlike traditional semantic or vector search engines—which make it difficult to count, filter, or chart large sets of matched documents—our new approach prioritizes transparency and performance while preserving semantic relevance.
Why we moved away from vector search
Our original implementation relied on semantic and vector search to capture the “meaning” behind user queries. But as our platform evolved, it became clear that these systems weren’t well-suited for our core use cases.
Users needed:
- Deterministic filtering (e.g., "how many results match this atom?")
- Transparent, complete result sets to power charts and dashboards
- Fast, repeatable queries that don’t change subtly over time
Modern vector search systems don’t easily support this level of transparency. They return approximate matches and abstract similarity scores, often making it hard to understand why a document was returned—or whether it’s the full picture.
So we made a decision: move away from vector search and lean into what traditional search engines do best.
A return to boolean and lexical search—with a twist
We rebuilt our search infrastructure on top of Elasticsearch’s powerful boolean and lexical search capabilities. This shift brings major advantages:
- Faster query speeds that dramatically improve iteration time
- Deterministic filtering and counts, so every chart is grounded in the full dataset
- Predictable, explainable results that users can trust
But we didn’t stop there.
To preserve the benefits of semantic understanding, we’ve rethought where that intelligence should live—not at query time, but at data ingestion.
Capturing semantic meaning at ingest time
Instead of computing document-query similarity during search, we enrich documents at the time of ingestion. Here’s how:
- Synonym expansion: We find related words and concepts not explicitly mentioned in the document and add them as fields, enabling semantic-style recall via lexical search.
- Stemming: Both queries and documents are reduced to their root forms, allowing consistent matches (e.g., “running” and “run”).
The result? You get the same functionality—semantically relevant results—without the opacity or latency tradeoffs of vector search.
What’s next: Reranking for even better relevance
We’re not done. Coming soon to Cypris is a reranking layer that boosts the most relevant results to the top of the list using lightweight vector techniques.
Here’s how it works:
- A standard lexical search retrieves the full result set.
- We take the top N results and rerank them using vector similarity, powered by Elasticsearch’s new hybrid scoring capabilities.
- You get faster queries with even better relevance—without compromising on counts or transparency.
This layered approach gives us the best of both worlds: precise filtering and fast queries, plus smarter ordering of results where it matters most.
We’re excited to bring this upgrade to our users, and we’re already seeing teams iterate faster and uncover insights more confidently. This is a foundational shift—and just the beginning of what’s to come.
Want a walkthrough of what’s changed? Reach out to our team.
Keep Reading
Neuromorphic computing is emerging as a distinct answer to the energy cost of artificial intelligence, and its patent landscape is unusually cross-disciplinary because a neuromorphic system is built from semiconductors, novel materials, and AI at the same time. Where conventional processors shuttle data between separate memory and compute units, an arrangement whose data movement dominates the energy budget, neuromorphic designs borrow from the brain: they compute where the data sits, in analog crossbar arrays that perform multiply-accumulate operations in place,¹,² communicate through sparse, event-driven spikes rather than continuous clocked operations, and store synaptic weights in analog or non-volatile devices.³ The intellectual property divides across several regions, each with different owners and maturity: the synaptic device materials, such as resistive, phase-change, and ferroelectric elements, that hold and update weights; the in-memory and analog compute circuits, often built as crossbar arrays, that perform computation in place; the spiking-processor architectures that route events across many cores; the event-based sensors, such as dynamic vision sensors, that feed them; and the on-chip learning rules and software stacks that make the hardware usable. Because a working system depends on all of these, freedom-to-operate and white space analysis must span the full stack.
The convergence of in-memory computing with spiking neural networks is now a well-reviewed field, spanning resistive, phase-change, ferroelectric, floating-gate, and optoelectronic synaptic devices,⁴,⁵ and it draws in several industries at once, which shapes where the IP concentrates. Large processor and memory companies, specialized neuromorphic startups, sensor makers, and academic groups are each building in different layers, so ownership is fragmented across the device, circuit, architecture, sensor, and algorithm regions rather than held by a single set of players. The patent record reflects this: across the Cypris corpus of more than 500 million patents and scientific papers, the neuromorphic, in-memory, and resistive-switching space holds on the order of 40,000 de-duplicated families and has grown steadily with a step-up in 2025, and the most active assignees are semiconductor and IT majors, including IBM, Hewlett Packard Enterprise, Samsung, and Intel, alongside strong academic filers, with China and the United States the leading jurisdictions; because assignee names are not fully canonicalized, corporate totals are best read as indicative. The commercial pull is strongest at the edge, where power and latency budgets are tight and brain-inspired efficiency has the clearest advantage. Because applications publish about eighteen months after filing, the most recent device and architecture filings are under-represented, so the current frontier is more active than granted-patent counts suggest.
The strategic question is which layer to back, and the white space sits where the physics is hardest. Analog in-memory computation promises the largest efficiency gains but must overcome device variability and precision limits, so materials and circuit techniques that make it reliable carry high, defensible value. Novel synaptic device materials, including optoelectronic elements that couple light and memory,⁷ and on-chip learning rules such as spike-timing-dependent plasticity demonstrated directly in memristor synapses,⁶ are active and comparatively open, while the software and compilation layers that connect neuromorphic hardware to mainstream AI frameworks remain underdeveloped and strategically important. Reading the landscape by layer, and tracking both the patents and the underlying device and algorithm research, is what separates a crowded region from an open one.
Where the neuromorphic white space is
Analog in-memory compute. Reliable analog computation in memory arrays promises the largest efficiency gains but must solve device variability and precision, a high-value, still-open target.¹
Novel synaptic devices. Resistive, phase-change, ferroelectric, and optoelectronic elements that store and update weights are an active materials layer with room for defensible positions.⁷
On-chip learning. Learning rules such as spike-timing-dependent plasticity that let a device adapt without a separate training system are a differentiated and comparatively open capability.⁶
Event-based sensing. Dynamic vision and other event-driven sensors that pair naturally with spiking processors are an active, less-crowded hardware layer.
Software and compilation stacks. Toolchains that map mainstream AI models onto neuromorphic hardware are underdeveloped and strategically important.
How AI-powered landscape and white space analysis helps
Resolving a landscape that spans device materials, compute circuits, processor architectures, sensors, and software requires more than keyword search. AI-powered analysis addresses this with semantic search that clusters activity by layer across varied terminology, attribution that normalizes semiconductor, startup, and academic filers to canonical entities, and continuous monitoring that keeps pace with a cross-disciplinary field. Because neuromorphic advances appear in scientific literature before they are patented, and because the field draws on materials, circuits, and AI at once, reading both patents and literature gives the earliest and fullest signal of where the frontier is moving.
Where Cypris fits
Cypris runs patent landscape and white space analysis for cross-disciplinary deep-tech fields such as neuromorphic computing across a corpus of more than 500 million patents and scientific papers, organized through a proprietary R&D ontology. The ontology clusters activity by layer, synaptic device, in-memory circuit, spiking architecture, event-based sensor, and learning and software, and normalizes semiconductor, startup, and academic filers to canonical entities, so a team can resolve which layers are crowded and which remain open as white space. Semantic search across patents and scientific literature connects filings to the underlying device and machine-learning research, which is where neuromorphic advances appear first, spanning the semiconductor, materials, and AI disciplines the field draws on. Cypris Q, the platform's agentic layer, lets teams run landscape and white space analysis conversationally and chain the clustering, attribution, and gap analysis, and Agentic Monitoring tracks a defined layer over time and flags new patents and papers as they publish. Cypris provides enterprise API partnerships with OpenAI, Anthropic, and Google, and is built with enterprise-grade security. Cypris serves hundreds of enterprise customers across pharmaceuticals, chemicals, advanced materials, energy, and other regulated industries.
FAQ
What is neuromorphic computing? Neuromorphic computing is a brain-inspired approach that processes information where it is stored, communicates through sparse event-driven spikes, and holds weights in analog or non-volatile devices, aiming to cut the energy cost of AI. It contrasts with conventional processors that separate memory and compute. Its advantage is clearest for low-power, low-latency workloads at the edge.
What layers does the neuromorphic patent landscape cover? The neuromorphic landscape covers synaptic device materials, in-memory and analog compute circuits, spiking-processor architectures, event-based sensors, and on-chip learning and software stacks. It is cross-disciplinary, drawing on semiconductors, materials, and AI. Freedom-to-operate and white space analysis must span all of these layers.
Who is active in neuromorphic computing patents? Activity spans large processor and memory companies, specialized neuromorphic startups, sensor makers, and academic groups, each building in different layers, so ownership is fragmented across the device, circuit, architecture, sensor, and algorithm regions. No single set of players holds the whole stack. That fragmentation makes structured landscape analysis valuable.
Where is the white space in neuromorphic computing? The white space includes reliable analog in-memory compute, novel synaptic device materials, on-chip learning, event-based sensing, and software and compilation stacks. Analog in-memory computation offers the largest efficiency gains but is the hardest to make reliable. The software layer that connects neuromorphic hardware to mainstream AI is underdeveloped and strategically important.
Why is in-memory compute a key IP area? In-memory compute is a key IP area because performing computation where data is stored avoids the energy cost of moving data, which is the main efficiency advantage of neuromorphic systems. Making analog in-memory computation reliable requires solving device variability and precision. The materials and circuit techniques that achieve this are foundational and defensible.
Why does neuromorphic analysis need scientific literature? Neuromorphic analysis needs scientific literature because device, circuit, and algorithm advances appear in research before they are patented, and the field's cross-disciplinary nature means relevant work spans several areas, so the literature gives the earliest and fullest signal. Analyzing patents alone gives a lagging, partial view. Cypris analyzes both across more than 500 million patents and scientific papers.
What software helps analyze the neuromorphic computing patent landscape? Software for the neuromorphic landscape should cluster activity by device, circuit, architecture, sensor, and software layer, resolve semiconductor, startup, and academic filers to canonical owners, search patents and scientific literature semantically, and monitor a cross-disciplinary field continuously. Cypris does this across more than 500 million patents and scientific papers using a proprietary R&D ontology, semantic search, Cypris Q, and Agentic Monitoring.
Which teams use neuromorphic patent landscape analysis? Neuromorphic patent landscape analysis is used by R&D, IP, and strategy teams at semiconductor, AI-hardware, and sensor companies, edge-AI developers, and their suppliers, as well as investors and research institutions. It informs which layer to back, where to file, and where competitors are concentrated. Cypris serves hundreds of enterprise customers across research-intensive and regulated industries.
Endnotes
- Musisi-Nkambwe, M., Afshari, S., Sanchez Esqueda, I., Kozicki, M. N., & Barnaby, H. (2021). The viability of analog-based accelerators for neuromorphic computing: a survey. Neuromorphic Computing and Engineering, 1(1). https://doi.org/10.1088/2634-4386/ac0242
- Hu, M., Rose, G. S., Chen, Y., Li, H., et al. (2014). Memristor crossbar-based neuromorphic computing system: a case study. IEEE Transactions on Neural Networks and Learning Systems, 25(10). https://doi.org/10.1109/tnnls.2013.2296777
- Xiao, Z., Hu, Q., Chu, P. K., Zhang, X., & Huang, A. (2018). Neuromorphic computing with memristor crossbar. Physica Status Solidi (a), 215(20). https://doi.org/10.1002/pssa.201700875
- Review of memristors for in-memory computing and spiking neural networks. (2025). Advanced Intelligent Systems. https://doi.org/10.1002/aisy.202500806
- Basu, A., & Hasler, J. (2024). Historical perspective and opportunity for computing in memory using floating-gate and resistive non-volatile computing including neuromorphic computing. Neuromorphic Computing and Engineering, 4(4). https://doi.org/10.1088/2634-4386/ad9b4a
- Pahlavan, S., Linares-Barranco, B., Serrano-Gotarredona, T., & Shooshtari, M. (2025). Spike-timing-dependent plasticity and synaptic consolidation in HfO2 memristors for adaptive neuromorphic computing. Neuromorphic Computing and Engineering. https://doi.org/10.1088/2634-4386/ae1da1
- Pereira, M., Kiazadeh, A., Martins, R., Fortunato, E., & Barquinha, P. (2023). Recent progress in optoelectronic memristors for neuromorphic and in-memory computation. Neuromorphic Computing and Engineering, 3(2). https://doi.org/10.1088/2634-4386/acd4e2
Post-quantum cryptography has moved from a research program to a mandated migration, and its patent landscape is distinctive because the value has shifted from the algorithms themselves to how they are implemented and deployed. A sufficiently powerful quantum computer would break the public-key cryptography, based on integer factorization and elliptic curves, that secures most digital communication today, and to prepare for that, the US National Institute of Standards and Technology finalized its first post-quantum standards, FIPS 203 (ML-KEM, for key establishment), FIPS 204 (ML-DSA), and FIPS 205 (SLH-DSA, for signatures), in August 2024, selected HQC as a fifth, backup key-establishment algorithm in 2025, and continues to develop further signature standards.⁷ The intellectual property divides across several regions, each a distinct area of patenting: the algorithm implementations across the lattice, hash, and code-based families; the hardware accelerators that make these computationally heavier algorithms fast enough for real systems;¹,⁵ the side-channel countermeasures that protect implementations from physical attack;²,³ the crypto-agility and migration tooling that let organizations discover and swap cryptography; and the integration of post-quantum schemes into protocols such as transport-layer security and into hardware roots of trust. Because a deployed system depends on several of these layers, freedom-to-operate and white space analysis must span the algorithm families and the implementation layers together.
The landscape has an unusual structure because of how the standards were set. NIST's standardization process operates under a patent-claim assurance framework: for any essential patent claim, the holder must either disclaim it or make a license available on reasonable-and-non-discriminatory or royalty-free terms, and patent questions around the leading lattice scheme were resolved through such licensing arrangements before finalization, so the foundational algorithm layer is comparatively open, though it is not accurate to call it "patent-free."⁸ That has pushed proprietary activity outward, toward the implementations and the migration ecosystem, where patenting is active and growing. The migration itself is not optional: NIST's draft transition guidance would deprecate the vulnerable classical algorithms after 2030 and disallow them after 2035, and national-security policy sets a 2035 migration target, while the "harvest-now, decrypt-later" threat, in which encrypted data captured today could be decrypted by a future quantum computer, gives the transition urgency even before large quantum computers exist.⁸ This is reshaping the record: across the Cypris corpus of more than 500 million patents and scientific papers, the post-quantum-cryptography set holds on the order of 4,642 families and rose from about 141 in 2020 to roughly 606 in 2024 and about 1,425 in 2025 on a partial count, an inflection that coincides with the standards' finalization, with the most active assignees a mix of chipmakers, banks, and platform vendors, including Intel, Wells Fargo, Huazhong University of Science and Technology, Huawei, IBM, and Samsung, and China ahead of the United States on geography; 2025 and 2026 counts are partial because of the publication lag.
The strategic question is which implementation layer to own, and the white space sits where the standardized algorithms meet real systems. Hardware acceleration for the lattice arithmetic and sampling that these algorithms require is a high-value layer, especially for constrained and Internet-of-Things devices where compute and power are limited.¹,⁵ Side-channel-resistant implementations are a distinct and heavily engineered layer, because a mathematically secure algorithm can still leak its keys through physical measurement, and even masked hardware implementations remain a target of attack research, so higher-order protection is an active frontier.²,³,⁴,⁶ Crypto-agility, the ability to inventory and swap cryptographic primitives across large systems, and migration tooling are a fast-growing ecosystem layer, as are hybrid schemes that run classical and post-quantum cryptography together during the transition, an option NIST accommodates rather than requires.⁸ Reading the landscape by algorithm family and implementation layer, and tracking both the patents and the underlying cryptography research, is what separates a crowded region from an open one.
Where the PQC white space is
Hardware acceleration. Accelerators for lattice arithmetic and sampling, especially for constrained and Internet-of-Things devices, are a high-value layer as the algorithms are computationally heavier than their predecessors.¹,⁵
Side-channel countermeasures. Implementations that resist physical attacks, which can leak keys even from a mathematically secure algorithm, are a distinct, heavily engineered layer where masking and higher-order protection are active frontiers.²,³,⁴,⁶
Crypto-agility and migration tooling. Discovering cryptographic assets across large systems and swapping primitives cleanly is a fast-growing ecosystem layer driven by migration deadlines.
Hybrid classical-and-post-quantum schemes. Running classical and post-quantum cryptography together during the transition is an active layer, particularly in protocols such as transport-layer security.
Protocol and root-of-trust integration. Embedding post-quantum schemes into protocols, secure elements, and hardware roots of trust is where deployment is decided.
How AI-powered landscape and white space analysis helps
Resolving a landscape that spans several algorithm families and implementation layers, under migration deadlines, requires more than keyword search. AI-powered analysis addresses this with semantic search that clusters activity by algorithm family and implementation layer across varied terminology, attribution that normalizes vendor, academic, and standards-linked filers to canonical entities, and continuous monitoring that keeps pace with a deadline-driven field. Because cryptography advances appear in scientific and conference literature before they are patented, reading both patents and literature gives the earliest signal of where the frontier and the white space are moving.
Where Cypris fits
Cypris runs patent landscape and white space analysis for fast-moving deep-tech fields such as post-quantum cryptography across a corpus of more than 500 million patents and scientific papers, organized through a proprietary R&D ontology. The ontology clusters activity by algorithm family, lattice, hash, and code-based, and by implementation layer, hardware acceleration, side-channel defense, crypto-agility, and protocol integration, and normalizes filers to canonical entities, so a team can resolve which families and layers are crowded and which remain open as white space. Semantic search across patents and scientific literature connects filings to the underlying cryptography research, which is where post-quantum advances appear first, often in preprints and conference proceedings ahead of the patent record. Cypris Q, the platform's agentic layer, lets teams run landscape and white space analysis conversationally and chain the clustering, attribution, and gap analysis, and Agentic Monitoring tracks a defined layer over time and flags new patents and papers as they publish. Cypris provides enterprise API partnerships with OpenAI, Anthropic, and Google, and is built with enterprise-grade security. Cypris serves hundreds of enterprise customers across pharmaceuticals, chemicals, advanced materials, energy, and other regulated industries.
FAQ
What is post-quantum cryptography? Post-quantum cryptography is a set of cryptographic algorithms designed to resist attack by quantum computers, which would break the public-key cryptography that secures most digital communication today. NIST finalized the first standards, mainly lattice-based schemes plus a hash-based signature scheme, in August 2024 and added a backup key-establishment algorithm in 2025. It is now moving into mandated deployment.
Why is PQC patenting shifting to implementations? PQC patenting is shifting to implementations because the core standardized algorithms are published under NIST's royalty-free or reasonable-and-non-discriminatory licensing-assurance framework, with the leading lattice scheme's patent questions resolved before finalization, leaving the algorithm layer comparatively open. Proprietary activity has therefore moved to hardware acceleration, side-channel defenses, crypto-agility, and protocol integration. That is where the growing patent activity now concentrates.
Are the post-quantum standards patent-free? No. The standards are published under NIST's patent-claim assurance framework, under which any essential patent claim must be disclaimed or licensed on royalty-free or reasonable-and-non-discriminatory terms, and specific licensing arrangements resolved the questions around the leading lattice scheme before finalization. That makes the algorithm layer comparatively open, but implementations, accelerators, and countermeasures are actively patented. "Comparatively open" is accurate; "patent-free" is not.
Why is migration to PQC urgent if quantum computers are not here yet? Migration is urgent because of the "harvest-now, decrypt-later" threat: an adversary can record encrypted data today and decrypt it once a capable quantum computer exists. NIST's draft transition guidance would deprecate vulnerable classical algorithms after 2030 and disallow them after 2035, and national-security policy sets a 2035 target. Long data lifetimes and slow cryptographic transitions make early action necessary.
Where is the white space in post-quantum cryptography? The white space includes hardware acceleration, especially for constrained and Internet-of-Things devices, side-channel countermeasures, crypto-agility and migration tooling, hybrid classical-and-post-quantum schemes, and protocol and root-of-trust integration. The standardized algorithms themselves are comparatively open. The higher-value opportunities are in the implementation and migration layers.
Why does PQC analysis need scientific literature? PQC analysis needs scientific literature because cryptography advances appear in research, preprints, and conference proceedings before they are patented, so the literature gives the earliest signal. Analyzing patents alone gives a lagging view. Cypris analyzes both across more than 500 million patents and scientific papers.
What software helps analyze the post-quantum cryptography patent landscape? Software for the PQC landscape should cluster activity by algorithm family and implementation layer, resolve vendor, academic, and standards-linked filers to canonical owners, search patents and scientific literature semantically, and monitor a deadline-driven field continuously. Cypris does this across more than 500 million patents and scientific papers using a proprietary R&D ontology, semantic search, Cypris Q, and Agentic Monitoring.
Which teams use PQC patent landscape analysis? Post-quantum cryptography patent landscape analysis is used by R&D, IP, and strategy teams at cybersecurity, semiconductor, cloud, and hardware-security companies, as well as investors and government-facing vendors. Because value concentrates in implementation layers under migration deadlines, structured analysis is essential. Cypris serves hundreds of enterprise customers across research-intensive and regulated industries.
Endnotes
- Xing, Y., & Li, S. (2021). A compact hardware implementation of CCA-secure key exchange mechanism CRYSTALS-KYBER on FPGA. IACR Transactions on Cryptographic Hardware and Embedded Systems, 2021(2). https://doi.org/10.46586/tches.v2021.i2.328-356
- Jati, A., Gupta, N., Chattopadhyay, A., & Sanadhya, S. K. (2023). A configurable CRYSTALS-Kyber hardware implementation with side-channel protection. ACM Transactions on Embedded Computing Systems, 22(2). https://doi.org/10.1145/3587037
- Mujdei, C., Beckers, A., Karmakar, A., et al. (2022). Side-channel analysis of lattice-based post-quantum cryptography: exploiting polynomial multiplication. ACM Transactions on Embedded Computing Systems. https://doi.org/10.1145/3569420
- Cabrera Aldaya, A., Camacho-Ruiz, E., & Navarro-Torrero, P. (2026). A framework for designing high-order side-channel-protected hardware implementations of ML-KEM (HOPE-MLKEM). IACR Transactions on Cryptographic Hardware and Embedded Systems, 2026(2). https://doi.org/10.46586/tches.v2026.i2.272-295
- Zhang, C., Zhang, Y., Wang, W., & Gu, D. (2024). Optimized hardware-software co-design for Kyber and Dilithium on RISC-V SoC FPGA. IACR Transactions on Cryptographic Hardware and Embedded Systems, 2024(3). https://doi.org/10.46586/tches.v2024.i3.99-135
- Ji, Y., & Dubrova, E. (2025). A side-channel attack on a masked hardware implementation of CRYSTALS-Kyber. Journal of Cryptographic Engineering, 15. https://doi.org/10.1007/s13389-025-00375-7
- National Institute of Standards and Technology. Post-quantum cryptography standardization (FIPS 203, 204, 205 finalized August 2024; HQC selected 2025). https://csrc.nist.gov/projects/post-quantum-cryptography/post-quantum-cryptography-standardization
- National Institute of Standards and Technology (2024). Transition to post-quantum cryptography standards (NIST IR 8547, initial public draft). https://csrc.nist.gov/pubs/ir/8547/ipd
Teams evaluating Clarivate's Cortellis for reaction and synthesis discovery are usually weighing a decades-old strength against a modern constraint. Cortellis is deep, trusted, and thorough. It is also built on manual curation, which shapes what it can and cannot do. Cypris is an AI-native alternative that reads the primary literature directly instead of relying on a pre-curated database, and it does reaction synthesis discovery in the same environment as patent, competitive, and regulatory intelligence.
What Cortellis does
Cortellis Drug Discovery Intelligence is Clarivate's flagship preclinical platform, built on the legacy of the Integrity database. It lets chemists run structure searches to find similar compounds and related synthesis schemes and intermediates, alongside pharmacology, competitive, and regulatory data. Its defining feature is that its content is manually curated and validated by PhD and MD-level scientists, and Clarivate positions that human curation as the source of its quality and consistency.
That curation is a real strength. It is also the constraint that leads teams to look for an alternative.
Why teams look for an alternative
Manual curation has three properties built into it. It is slow, because a person reads each source. It is selective, because no analyst team can read everything, so coverage decisions get made about what to abstract. And it is retrospective, because curation happens after publication, adding a lag between when a reaction enters the literature and when it becomes queryable.
For reaction synthesis discovery, those compound. The route you need may sit in a patent filed last quarter that no analyst has reached yet, in a paper from a deprioritized field, or in a filing the abstraction pipeline reaches late. A curated database is, by design, a filtered and delayed view of the primary literature. For most of the last thirty years that was the best available option. It no longer is.
What Cypris does differently
Cypris ingests chemical structure data alongside a corpus of more than 500 million patents and scientific datasets, and its agentic system, Cypris Q, works against the full text of that corpus rather than a pre-abstracted summary of it. Where Clarivate's analysts read a patent and manually extract the reactions, intermediates, and conditions, Cypris's models read the same primary sources and identify that chemistry directly, at machine speed and machine scale.
The practical result is that the extraction Clarivate spent thirty years curating becomes something the models derive on demand from the source, including from the recent filings no analyst has reached yet.
Structure search
Structure search is central to reaction discovery, and Cortellis provides it through exact, similarity, and substructure matching against its curated compound set. Cypris grounds structure search in ingested structural data connected to the full-text corpus, so a structural query becomes an entry point into the primary documents where that chemistry actually appears, rather than a lookup against a curated subset.
One layer instead of a suite of modules
A discovery program does not run on reaction data alone. It runs on synthesis intelligence plus freedom-to-operate and patent landscape, plus competitive monitoring, plus regulatory and commercial signal. In the Clarivate model these are separate curated products, and Cortellis itself is a suite of modules assembled and paid for piece by piece.
Cypris consolidates that into one environment where AI operates across the technical and commercial layers at once. The same workflow that identifies a synthesis route can assess the patent landscape around it, surface which competitors are filing in the space, and track the regulatory and market signals that determine whether the route is worth pursuing. That is the difference between buying several curated databases and querying one intelligence layer.
Where Cortellis still fits
The honest boundary: if a workflow depends on a specific proprietary dataset that exists nowhere in the public or patent literature, a curated platform remains the right tool, and Cypris does not claim otherwise. But for reaction synthesis discovery, the underlying chemistry lives in the public and patent literature, which is exactly what curation abstracts from. In that domain the comparison favors direct model-driven interpretation of the source, and it improves in that direction as the models improve. A curated database advances at the speed of its curation team. An AI-native layer advances at the speed of its models.
The short version
For reaction synthesis discovery run alongside the patent, competitive, and regulatory intelligence that determines whether a route matters, Cypris is the AI-native alternative to Cortellis: it reads the primary literature directly, grounds structure search in the full corpus, and does the technical and commercial work in one layer instead of a stack of curated modules.
FAQ
Is Cypris a direct alternative to Clarivate Cortellis?
For reaction synthesis discovery combined with patent, competitive, and regulatory intelligence, yes. Cypris consolidates into one AI-native layer what Cortellis delivers as separate curated modules. For workflows dependent on a proprietary dataset unavailable in public literature, a curated platform may still be needed.
What is the core difference between Cypris and Cortellis?
Data model. Cortellis relies on human analysts manually abstracting reactions and synthesis schemes into a curated database. Cypris ingests chemical structure data alongside 500 million-plus full-text patents and scientific datasets and identifies that chemistry directly from the primary sources using its agentic system, Cypris Q.
Does Cypris support chemical structure search?
Yes. Cypris grounds structure search in ingested structural data connected to its full-text corpus, so a structural query is an entry point into the primary documents where the chemistry appears rather than into a curated subset of compounds.
What does Cortellis do for reaction synthesis?
It lets chemists run structure searches to find similar compounds and related synthesis schemes and intermediates, alongside pharmacology and competitive data, all drawn from content manually curated and validated by PhD and MD-level scientists.
Why would a team move off a curated database?
Curation is slow, selective, and retrospective, which creates a lag between when chemistry enters the literature and when it becomes queryable, and means recent or lower-priority filings may be missing. Reading the primary corpus directly removes that lag.
Is manual curation still valuable?
For datasets that exist nowhere in public or patent literature, yes. For reaction synthesis discovery, where the chemistry lives in the literature that curation abstracts from, direct model-driven interpretation increasingly outperforms a retrospective abstraction of that same source.
How does Cypris handle recent filings better?
Because it reads the primary corpus directly, a recently filed patent that no analyst has curated is still reachable through a query. Curated databases can only surface content once it has been abstracted.
What does the "single layer" advantage mean in practice?
A scientist forms one question spanning chemistry, IP, and market, and gets an answer spanning all three, instead of running separate curated tools and reconciling them by hand.
Which teams is Cypris the better fit for?
Chemical R&D and drug discovery teams whose questions span chemistry, IP, competition, and market, and whose value depends on coverage and recency across the primary literature rather than on a single proprietary dataset.
What is Cypris Q?
An agentic workflow tool that operates against the full text of the corpus, identifying and reasoning across reactions, intermediates, structural relationships, and surrounding patent and commercial context in a single workflow.