Privacy-Enhancing Computation Patents: Who Leads, Where Gaps Are 2026
- Filing activity peaked in 2021 at 165 records and has since flattened, with the 2022 midpoint of 158 confirming the plateau rather than continued growth.
- H04L dominates at 1,116 of 1,214 records, while G06N's 178 records show AI-adjacent privacy computation is a smaller but distinct claim zone.
- Recent-year momentum has collapsed across almost every major assignee, several tracked leaders show 0 filings in the latest year, a -100% YoY drop from prior activity.
Filing growth compares 2021 (165 records) with 2024 (163) — a three-year span. 2024 is the most recent year we treat as complete: publication lags filing by roughly 18 months, so 2025 onwards are still filling in and any growth rate that ends there would understate the field. Top-5 share is the combined record count of the five largest assignees divided by all 1,214 records in scope (CR5), not by the ranked leaders only.
What the filing record shows
Privacy-enhancing computation spans homomorphic encryption, secure multiparty computation and related privacy-enhancing technologies, tied together in this dataset by claims around computation overhead, bootstrapping, secret sharing, threat models and practical deployment. The 1,214 patent families captured here span 2015 through the middle of 2026, with filings concentrated under H04L (digital information transmission) and a substantial secondary cluster in G06F (electric digital data processing).
Filing volume rose sharply through the late 2010s, peaked in 2021, and has been flat-to-declining since, a pattern consistent with a field that built out its core claim space early and has not yet found a second wave of novel filing activity. Publication lag means the last one to two years understate true filing activity, but the plateau visible from 2021 onward predates that lag window and is likely a genuine slowdown rather than a reporting artefact.
Filing trends and technology composition
Two views of the same 1,214-family dataset: the year-by-year filing curve, and the IPC subclasses where those filings actually sit.
A decade of growth, then a plateau
From 37 filings in 2017 to a peak of 165 in 2021, the field grew roughly fourfold in four years. Filing counts since then have held near that level rather than continuing to climb, and the most recent partial year should be read with the 18-month publication lag in mind rather than as a sign of collapse.
H04L carries the claim weight
H04L accounts for 1,116 of 1,214 records, confirming that most privacy-enhancing computation claims are framed as digital-transmission security rather than general-purpose data processing. G06F's 476 records and G06N's 178 show meaningful secondary activity in data-processing and AI-model contexts, while G16H (healthcare informatics, 33), G09C (ciphering, 28) and G16Z (specific-application computing, 15) remain comparatively thin.
Shares are the percentage of the 1,214 records in scope. A patent can carry several IPC classes, so the shares add up to more than 100%.
Go deeper on Privacy-Enhancing Computation with Eureka
This page is one run against one query. Ask Eureka your own question about privacy-enhancing computation and every answer comes back with the patent numbers behind it.
Try EurekaThe most-cited filings in this space
Distributed registration and authentication via threshold secret sharing and additively homomorphic encryption
Filed by VMWARE LLC and published 2023-10-26, this filing combines a threshold secret sharing scheme, where reconstructing a secret requires cooperation from at least T+1 of N parties, with additively homomorphic encryption that lets computations run on encrypted data without prior decryption. Together they support distributed node registration and authentication without any single party holding the full secret.Abstract condensed from the original filing.
| # | Publication no. | Patent title | Citations |
|---|---|---|---|
| 1 | US20190036678A1 | Systems and methods for implementing an efficient, scalable homomorphic transformation of encrypted data with… | 422 |
| 2 | US20110110525A1 | Fully homomorphic encryption method based on a bootstrappable encryption scheme, computer program and apparat… | 266 |
| 3 | US20130170640A1 | Fully Homomorphic Encryption | 182 |
| 4 | US20160105402A1 | Homomorphic encryption in a healthcare network environment, system and methods | 165 |
| 5 | US20180276417A1 | Secure multiparty computation on spreadsheets | 163 |
| 6 | US20190386814A1 | Systems and Methods for Implementing an Efficient, Scalable Homomorphic Transformation of Encrypted Data with… | 150 |
| 7 | US20190334716A1 | Blockchain-empowered crowdsourced computing system | 149 |
| 8 | US20160119119A1 | Compact fuzzy private matching using a fully-homomorphic encryption scheme | 117 |
| 9 | US20140177828A1 | Managed secure computations on encrypted data | 117 |
| 10 | US20200151356A1 | System and method for fast and efficient searching of encrypted ciphertexts | 103 |
Citation counts favour older filings that have had more time to accumulate references; treat them as a signal of influence within this corpus, not of current commercial relevance.
Each row carries its publication number; clicking a row searches Eureka by that number.
Put your own technology through the same analysis
Eureka on the web
When you want the answer in the next five minutes.
The agent works the prompt against patents and technical literature, citing every source.
Run your analysis now →MCP server & REST API
When it has to run inside your own pipeline.
Patent search, landscape analysis and assignee resolution as MCP tools. Drop them into any agent framework, or call REST directly.
Browse MCP servers →What the data means for filing strategy
Three patterns stand out once the raw counts are read against filing dates and IPC placement.
The core wave has already broken
Growth from 37 filings in 2017 to a 2021 peak of 165 reflects the initial land rush around homomorphic encryption and secure multiparty computation. The 2022 count of 158 and the subsequent flat trajectory suggest the obvious claim territory in this framing is largely staked out.
Transmission security still frames most claims
Nearly all filings sit under H04L, meaning the dominant framing is protecting data in transit or in communication protocols rather than general-purpose secure computation. The G06F cluster is large enough to matter but is a clear secondary track, not a rival center of gravity.
Even top filers have gone quiet
Several of the most active historical assignees, including major electronics and academic-industry consortia, show zero filings in the latest tracked year. This is consistent with the broader plateau rather than any single company exiting the field.
US and China lead, but PCT and EPO matter
The United States and China together account for the largest share of receiving-office activity, with WIPO PCT filings (130) and EPO filings (135) indicating that a meaningful portion of applicants are still pursuing multi-jurisdiction protection rather than filing domestically only.
Eureka can read the same corpus for gaps instead of for coverage: under-claimed branches adjacent to privacy-enhancing computation, with the prior art for and against each one.
Who is filing, and who has stopped
Co-assignee pairs point to a small number of tightly linked filing clusters, mostly built around university-industry technology transfer arrangements rather than pure corporate R&D.
Huawei's filing pace has cooled sharply
Huawei remains one of the more active corporate filers historically in this space but shows only a single filing in the latest tracked year, an 83% year-on-year decline consistent with the broader plateau.
Samsung and SNU R&DB Foundation form the densest cluster
The strongest co-assignee pairing links Samsung Electronics with SNU R&DB Foundation at 38 shared records, with a related pairing through Korea University's Industry-Academic Cooperation Foundation adding two more dense links. This points to a concentrated Korean university-industry filing network.
IBM's filing activity has also stalled
IBM appears among the historically significant assignees but registers no filings in the most recent tracked year, mirroring the pattern seen across Samsung, SNU R&DB Foundation and Zama SAS.
| Assignee | Recent year | YoY |
|---|---|---|
| Huawei Technologies Co., Ltd. | 1 | -83% |
| Samsung Electronics Co., Ltd. | 0 | -100% |
| International Business Machines Corporation (IBM) | 0 | — |
| SNU R&DB Foundation | 0 | -100% |
| ZAMA SAS | 0 | -100% |
| Korea University Industry-Academic Cooperation Foundation | 0 | — |
| Microsoft Technology Licensing, LLC | 0 | — |
| Thales DIS France SA | 0 | — |
Where to take this
The filing plateau does not mean the technology is settled; it means the obvious claims are filed and the remaining opportunity is narrower and more specific.
Check freedom-to-operate against the top-cited cluster
The most-cited records in this dataset, several tied to foundational fully homomorphic encryption schemes, sit on dense prior art. Any new filing touching bootstrapping or ciphertext expansion should be checked against these first.
Run a citation check in EurekaWatch the Korean university-industry cluster
The strongest co-assignee pairings all involve Korean academic technology-transfer foundations working alongside Samsung. Their next filings are a reasonable leading indicator for where applied homomorphic encryption research is heading.
Track assignee activity in EurekaTarget the under-claimed sub-areas directly
Bootstrapping latency, threshold key rotation and edge-device threat modelling all show thinner filing density than the core transmission-security cluster, making them more realistic targets for a defensible new claim.
Explore white space in EurekaCommon questions about this landscape
In this dataset it means filings whose title or abstract reference homomorphic encryption, secure multiparty computation or privacy-enhancing technology, combined with claim language around computation overhead, bootstrapping, secret sharing, threat models or practical deployment. The IPC scope is restricted to H04L9, G06F21 and G06F7, which covers cryptographic mechanisms and digital data processing but excludes purely hardware-level security. This is a targeted slice of the broader cybersecurity patent space, not every cryptography filing that mentions privacy.
The filing curve grew from 37 records in 2017 to a peak of 165 in 2021, then held near that level through 2022 at 158 before the trend flattened. This pattern typically appears when a field's foundational claim space, in this case core homomorphic encryption and secure multiparty computation constructions, has been substantially staked out by early filers. It does not mean research activity has stopped, only that the rate of new distinct patent claims has slowed. The most recent one to two years in any filing dataset are also understated because publication lags filing by roughly 18 months.
The dataset's co-assignee and momentum data point to a mix of large electronics manufacturers and Korean university technology-transfer foundations as the most historically active filers, with Samsung Electronics and SNU R&DB Foundation forming the strongest co-filing pair at 38 shared records. Huawei, IBM and Zama SAS also appear among the more active historical assignees. Notably, several of these same organisations show sharply reduced or zero filings in the latest tracked year, suggesting the most active period of filing has already passed for the current leaders.
This dataset does not split filing counts separately by homomorphic encryption versus secure multiparty computation, since the search terms combine both under one privacy-enhancing computation umbrella. What is clear from the IPC split is that H04L (digital transmission security) carries the overwhelming majority of claims at 1,116 of 1,214 records, meaning both techniques are predominantly framed and claimed as communication-security mechanisms rather than as general-purpose computation methods under G06F.
The clearest gaps sit outside the dominant H04L cluster: healthcare-specific applications (G16H) account for only 33 records, ciphering-specific claims (G09C) only 28, and specific-application computing (G16Z) just 15, all thin relative to the 1,116-record H04L core. Practically, that points toward bootstrapping latency reduction, threshold secret-sharing key rotation, and threat-model formalization for edge or IoT deployments as areas where filing density is lower and a well-drafted claim has more room to stand on its own.
Research Privacy-Enhancing Computation in depth with Eureka
Go past this page: query the whole privacy-enhancing computation corpus yourself, in your own scope.
Every answer comes back with patent numbers you can open.
Disclaimer. This page is generated from Patsnap Eureka data drawn from a limited snapshot of global patent and scientific-literature records, and is provided for general information and reference only.
Patent data carries inherent limitations: recent filings (typically the most recent 18–24 months) are under-counted due to standard publication lag; counts may be reported at either a patent-family or a patent-record basis and are not always directly comparable; classification, applicant-name, and citation data may contain errors, duplicates, or omissions; and the underlying search query defines and constrains the scope shown. As a result, the analysis may be incomplete or inaccurate and may not reflect the full technology landscape.
Nothing on this page constitutes an exhaustive prior-art, novelty, freedom-to-operate, or validity search, nor does it constitute legal, financial, investment, or professional advice, and it should not be relied upon as such. Any patent, commercial, or strategic decision should be verified independently and reviewed with qualified patent, legal, and domain professionals. Patsnap makes no warranties, express or implied, as to the accuracy, completeness, or fitness for any particular purpose of the information presented.
Machine translation. Assignee and organisation names originally recorded in Chinese, Japanese or Korean have been rendered into English by an AI translation step so that the tables stay readable. These renderings are best-effort and may not match a company’s registered English name; the original name is what the underlying patent record carries, and it is what any Eureka query launched from this page uses.