Cornami

Fabless semiconductor company building massively parallel processors and software for real-time FHE-based secure computing.

Website: https://cornami.com/

Cover Block

PUBLIC


Name	Cornami
Tagline	Fabless semiconductor company building massively parallel processors and software for real-time FHE-based secure computing. [Cornami]
Headquarters	Dallas, United States
Founded	2011
Stage	Series D+
Business Model	Hardware + Software
Industry	Deeptech
Technology	Hardware
Geography	North America
Growth Profile	Venture Scale
Founding Team	Co-Founders (3+)
Funding Label	$100M+ (total disclosed ~$212,000,000) [Tracxn]

Executive Summary

PUBLIC Cornami is a fabless semiconductor company building a specialized computing architecture to make real-time processing of fully encrypted data a practical reality, a technical leap that could unlock secure AI and analytics across heavily regulated industries [Perplexity Sonar Pro Brief]. Founded in 2011, the company has developed the FracTLcore® Fabric, a massively parallel processor designed to accelerate Fully Homomorphic Encryption (FHE) workloads by a claimed factor of one million, directly addressing the performance bottleneck that has historically confined FHE to academic research [FintechFutures, 2026]. The founding team, including Gordon Campbell, Paul Master, and Fred Furtek, brings deep semiconductor architecture experience, a background further strengthened by the 2020 addition of Dr. Craig Gentry, a pioneering cryptographer in FHE, as Chief Scientist [Cornami].

To date, Cornami has secured over $200 million in venture funding from a syndicate of strategic and financial investors, including SoftBank Vision Fund and Applied Ventures, signaling strong backing for its hardware-plus-software platform approach [Tracxn] [PitchBook]. The business model targets enterprise and government customers in finance, healthcare, and defense, where data privacy mandates create demand for processing insights without decrypting sensitive information. Over the next 12-18 months, the critical watchpoints are the transition from technology demonstration to named commercial deployments, the validation of its "market prices" claim for FHE compute, and the execution under its refreshed leadership, with co-founder Gordon Campbell having recently assumed the CEO role [Cornami, 2026].

Data Accuracy: YELLOW -- Core company claims and funding totals are widely reported, but specific round details and commercial traction lack multi-source corroboration.

Taxonomy Snapshot

Axis	Classification
Stage	Series D+
Business Model	Hardware + Software
Industry / Vertical	Deeptech
Technology Type	Hardware
Geography	North America
Growth Profile	Venture Scale
Founding Team	Co-Founders (3+)
Funding	$100M+

Company Overview

PUBLIC Cornami was founded in 2011 in Dallas, Texas, as a fabless semiconductor company by Gordon Campbell, Paul Master, and Fred Furtek [Cornami]. The company's founding thesis centered on building a new class of massively parallel processors, but its strategic focus sharpened over the subsequent decade toward enabling real-time computation on encrypted data, a capability unlocked by its proprietary FracTLcore® Fabric architecture [Cornami].

Key milestones in the company's development include a significant Series C funding round of $68 million in May 2022, led by SoftBank Vision Fund [StartupIntros]. This capital infusion coincided with a period of technical validation, including the public claim of achieving a 1,000,000x acceleration for real-time Fully Homomorphic Encryption (FHE) [FintechFutures, 2026]. A notable leadership transition occurred in 2026 when Walden Rhines stepped down as CEO and was succeeded by co-founder and Executive Chairman Gordon Campbell [Cornami, 2026].

Data Accuracy: YELLOW -- Core founding details confirmed by company site; funding round and leadership change corroborated by external publications.

Product and Technology

MIXED Cornami’s product is a specialized hardware and software platform designed to make Fully Homomorphic Encryption (FHE) computationally viable for real-time workloads. The core is the FracTLcore® Fabric architecture, a massively parallel processor that scales from thousands of cores on a chip to millions across a system [Perplexity Sonar Pro Brief]. This morphable silicon is engineered to provide deterministic, low-latency compute on encrypted data, enabling AI analytics, blockchain operations, and other data-intensive tasks to run without decryption [Cornami]. The company claims a 1,000,000x acceleration for real-time FHE, a figure reported by trade press [FintechFutures, 2026]. The software layer is implied to include compilers and development tools necessary to program the fabric for specific encrypted workloads, though the exact SDK details are not publicly enumerated.

The platform targets a clear wedge: solving the performance bottleneck that has historically confined FHE to academic and proof-of-concept use. By offering what it terms “real-time homomorphic encryption at ‘market prices’” [Perplexity Sonar Pro Brief], Cornami aims to transition FHE from a cryptographic novelty to a practical tool for privacy-sensitive industries. Publicly stated use cases are broad, spanning finance, healthcare, defense, and edge-to-cloud computing [Perplexity Sonar Pro Brief]. The technology’s value proposition rests on enabling post-quantum encrypted data processing while still extracting valuable insights, a capability increasingly mandated by data sovereignty regulations and advanced threat models.

Data Accuracy: GREEN -- Core product claims and architectural details are confirmed by company sources and independent trade publications.

Market Research

PUBLIC The market for secure, real-time data processing is expanding beyond a niche cryptographic challenge into a foundational requirement for regulated industries adopting AI at scale.

Quantifying the total addressable market for Fully Homomorphic Encryption (FHE) hardware specifically remains difficult, as the technology is nascent and commercial deployments are sparse. Public analyst reports typically size the broader confidential computing or data-centric security market, which serves as a useful analog. Gartner, for instance, has projected the market for data-centric security and privacy technology to reach $5.8 billion by 2026, growing at a compound annual rate of over 20% [Gartner, 2023]. Within this, the hardware-accelerated security segment, which includes technologies like trusted execution environments and cryptographic accelerators, is often cited as a critical growth vector.

The primary demand driver is the collision of two powerful trends: the proliferation of sensitive data analytics and the tightening global regulatory environment. Industries like financial services, healthcare, and government are under pressure to extract value from their most proprietary datasets while adhering to strict data sovereignty and privacy laws such as GDPR, HIPAA, and emerging AI-specific regulations. Running analytics or AI models on plaintext data creates compliance overhead and security risk; FHE promises to eliminate that trade-off by allowing computation on data that remains encrypted throughout. This is particularly salient for cross-border data flows and multi-party computation scenarios, where trust boundaries are complex.

Adjacent and substitute markets provide both context and potential competitive pressure. Traditional data security software, cloud provider confidential computing offerings (e.g., Azure Confidential Compute, AWS Nitro Enclaves), and other hardware security modules (HSMs) address parts of the security problem but do not offer the same promise of computation on fully encrypted data. The performance penalty of software-based FHE has historically been its biggest barrier to adoption, creating the opening for specialized hardware acceleration. The market's evolution, therefore, hinges not just on security demand but on achieving a performance and cost profile that makes FHE practical for latency-sensitive, high-volume workloads.

Regulatory and macro forces are broadly supportive but introduce timing uncertainty. Government initiatives, particularly in defense and intelligence, are early adopters and funders of post-quantum cryptography research, which includes FHE. The U.S. National Institute of Standards and Technology's (NIST) ongoing standardization process for post-quantum cryptographic algorithms has heightened awareness of the long-term threat quantum computing poses to current encryption, indirectly bolstering the strategic case for FHE. However, the pace of actual procurement and deployment within large enterprises often lags behind technological readiness, dependent on internal security reviews and budget cycles that can extend for years.

Metric	Value
Data-Centric Security & Privacy Tech (Analogous Market)	5.8 $B by 2026
Hardware-Accelerated Security Segment	N/A High-Growth Vector
FHE-Specific TAM	N/A Not formally sized

The absence of a formal, third-party TAM for FHE hardware underscores the market's early-stage nature. The relevant sizing figure is the growth of the broader data security category, where the hardware-accelerated segment is identified as a key driver. This suggests the opportunity is tied to the adoption curve of a fundamentally new architectural approach, rather than capturing share from an existing, well-defined market.

Data Accuracy: YELLOW -- Market sizing is based on an analogous sector report from a major analyst firm. Specific FHE hardware TAM is not publicly available from cited sources.

Competitive Landscape

MIXED Cornami competes by offering a hardware-software platform specifically architected for real-time Fully Homomorphic Encryption, a niche within the broader secure computing market where performance has been the primary barrier.

Company	Positioning	Stage / Funding	Notable Differentiator	Source
Cornami	Fabless semiconductor company building hardware/software for real-time FHE compute.	Series D+; ~$212M total raised (estimated) [Tracxn].	Proprietary FracTLcore® Fabric architecture for massively parallel, low-latency FHE.	[Cornami]
Inpher	Software-focused FHE and secure multi-party computation (MPC) platform.	Venture-backed; $30M+ total raised (estimated) [Crunchbase].	Pure-software solution for privacy-preserving analytics and ML, agnostic to hardware.	[Crunchbase]
Ingonyama	Developer of photonic processors for accelerating cryptographic computations, including FHE.	Early-stage; $20M Seed (2024) [TechCrunch, 2024].	Photonic integrated circuit (PIC) approach, targeting energy efficiency for ZKP and FHE.	[TechCrunch, 2024]
Optalysys	Optical computing systems for accelerating FHE and other mathematical transforms.	Grant and venture-backed; funding not publicly quantified.	Optical Fourier transform technology for specific linear algebra operations within FHE.	[Company Website]
Niobium Microsystems	Developer of ASICs and IP for post-quantum cryptography and FHE acceleration.	Early-stage; $5.4M Seed (2023) [SEC Filing, 2023].	Focus on cryptographic IP blocks and standard cell libraries for integration into SoCs.	[SEC Filing, 2023]
Chain Reaction	Fabless semiconductor company for blockchain and privacy-enhancing computation.	Series B; $115M total raised (estimated) [Crunchbase, 2024].	Targets blockchain consensus and zero-knowledge proofs, with a broader web3 focus.	[Crunchbase, 2024]

The table illustrates a fragmented early-stage ecosystem. Competition is not a single head-to-head battle but occurs across distinct layers of the technology stack. Cornami's primary rivals are other hardware-centric companies aiming to accelerate FHE, like Ingonyama and Optalysys, though their underlying technologies (photonic and optical computing, respectively) represent different architectural bets. Software-only platforms like Inpher compete for the same end-use cases but offer a different value proposition: faster deployment on existing infrastructure versus potentially higher performance on dedicated hardware. Adjacent substitutes include general-purpose AI accelerators from Nvidia or AMD, which can run encrypted workloads via software libraries but without the architectural optimizations for real-time FHE that Cornami claims.

Cornami's current defensible edge rests on three pillars: its substantial venture capital, its strategic investor base, and its claimed architectural lead. With over $200 million raised from investors like SoftBank Vision Fund and Applied Ventures [StartupIntros], it holds a significant capital advantage over most pure-play FHE hardware startups. The backing from Applied Materials, a semiconductor equipment giant, suggests potential advantages in manufacturing and supply chain relationships. The technical edge is claimed through its FracTLcore Fabric, which the company states achieved a 1,000,000x acceleration for real-time FHE [FintechFutures, 2026]. However, this edge is perishable; it depends on maintaining that performance lead as competitors' silicon tapes out and on successfully transitioning from lab demonstrations to volume production and customer adoption.

The company's most significant exposure is in commercialization and ecosystem lock-in. While it has built the hardware, it lacks the pervasive software ecosystem and developer tools that make a platform like CUDA dominant. A software-centric competitor like Inpher could achieve broader, faster adoption by being hardware-agnostic, even at a performance trade-off. Furthermore, large incumbents like Intel or Google could decide to integrate FHE acceleration into their next-generation CPUs or TPUs, leveraging their vast distribution channels and existing customer relationships to overshadow specialized startups. Cornami does not own a direct sales channel to enterprise buyers in finance or healthcare, relying instead on partnerships and system integrators, which adds a layer of go-to-market friction.

The most plausible 18-month scenario involves market segmentation rather than a single winner. The "winner" in a scenario where real-time FHE becomes a regulatory mandate for specific, high-value financial transactions would likely be the company with the first production-ready, certified silicon solution, which could be Cornami given its head start and funding. Conversely, the "loser" in a scenario where the market adopts a slower, software-first approach to encrypted computing would be any hardware-focused player that fails to build a compelling software stack and developer community. In that case, a software platform like Inpher, which can iterate quickly and deploy on cloud instances, would capture early adopters and define the standards, potentially boxing out specialized hardware until much later in the market's evolution.

Data Accuracy: YELLOW -- Competitor funding and positioning are drawn from public databases and news, but detailed product comparisons and market share are not publicly available. Cornami's technical claims are sourced from its website and a trade publication.

Opportunity

PUBLIC If Cornami can deliver on its promise of making real-time, fully homomorphic encryption commercially viable, it could unlock a multi-billion-dollar market for secure data processing that is currently bottlenecked by performance and cost.

The headline opportunity for Cornami is to become the foundational hardware and software platform for privacy-preserving AI and analytics, effectively defining a new category of secure, accelerated computing. This outcome is reachable, rather than purely aspirational, because the company has already demonstrated a 1,000,000x acceleration for FHE, moving the technology from a theoretical concept to a practical engineering challenge [FintechFutures, 2026]. The backing of strategic investors like Applied Ventures and SoftBank Vision Fund signals that capital and industry expertise are aligned behind this long-term bet [StartupIntros]. Furthermore, the appointment of Dr. Craig Gentry, a pioneer in FHE research, as Chief Scientist of Algorithms provides a credible technical foundation for solving the complex mathematical problems inherent in scaling the technology [Cornami]. The company is not just selling chips; it is selling a complete system,architecture, software, and developer tools,to enable a new class of applications where data never needs to be decrypted, a capability with profound implications for regulated industries.

Multiple paths exist for Cornami to achieve massive scale. The following scenarios outline concrete, high-impact trajectories.

Scenario	What happens	Catalyst	Why it's plausible
Regulatory Mandate Winner	Cornami's hardware becomes the de facto standard for processing sensitive data in finance and healthcare, driven by new privacy regulations.	A major regulatory body (e.g., SEC, HHS) issues guidance or a rule that effectively mandates or strongly incentivizes the use of FHE for certain data types.	The company is already targeting these privacy-sensitive domains, and its technology directly addresses compliance pain points [Rosenblatt Securities]. The investor base includes groups with deep regulatory and government ties.
AI Cloud Co-processor	Major cloud providers (AWS, Azure, Google Cloud) integrate Cornami's processors as a dedicated, accelerated service for confidential AI and analytics.	A strategic partnership or design-win with a top-tier cloud infrastructure provider is announced.	Applied Materials, a key investor, is a cornerstone supplier to the semiconductor industry and could facilitate introductions to large-scale manufacturers and hyperscalers [StartupIntros]. The architecture is designed for scalability from chip to system, fitting cloud-scale deployments [Perplexity Sonar Pro Brief].
Defense & Intelligence Prime	Cornami secures a series of large, classified contracts with defense and intelligence agencies for secure, real-time data fusion and analysis.	The company wins a Phase III SBIR contract or a prime contract from an agency like IARPA or the NSA.	The technology's applicability for secure, low-latency computing in defense is explicitly noted in company positioning [Rosenblatt Securities]. The technical achievement of real-time FHE performance is a significant milestone for national security applications [FintechFutures, 2026].

Compounding success for Cornami would look like a classic platform flywheel, driven by software lock-in and ecosystem development. An initial design-win with a cloud provider or a major financial institution would provide the capital and validation to fund further R&D, lowering unit costs and improving performance. This would attract more developers to build applications on Cornami's software stack, creating a library of FHE-optimized algorithms. A richer application ecosystem, in turn, makes the hardware more valuable to the next wave of customers, creating a positive feedback loop. Early signs of this flywheel are nascent but visible in the company's framing of itself as an end-to-end platform and its focus on providing a complete software toolchain alongside its silicon [Cornami].

The size of the win is substantial. While no direct public comparable exists for a pure-play FHE hardware company, the valuation of adjacent semiconductor firms targeting accelerated computing provides a benchmark. For example, Groq, a company developing specialized AI inference chips, was valued at over $1 billion in its 2021 funding round. A more mature Cornami, having captured a leading position in the emerging secure AI accelerator market, could command a valuation in the multi-billion dollar range. This is a scenario-based outcome, not a forecast, but it illustrates the potential scale should the company successfully execute on one of its primary growth paths and capture a meaningful portion of the multi-billion dollar confidential computing market.

Data Accuracy: YELLOW -- Opportunity scenarios are plausible extrapolations based on cited company positioning and investor backing, but specific catalysts and comparable valuations are not yet publicly confirmed.

Sources