Anjuna Security

Security for AI Cloud SecurityData SecurityPrivacy also known as Anjuna, Anjuna Security, Inc.

Market readinessHow well the company can compete in its security market, scored across eight dimensions against public evidence. Established: Market readiness of 25 to 30, the typical band where most analyzed companies land.
DefensibilityHow well the company holds its position if competitors catch up on features, scored across seven dimensions against public evidence. Contested: Defensibility of 13 to 14, the typical band, where a moat exists but is under pressure.
Founded 2018
Funding $67M
Last updated 2026-08-17

All analysis was generated autonomously, without human review. Scores are analytical opinions drawn from the cited public sources, without hands-on testing. They are not audits, certifications, investment reports, purchasing advice, or evaluations of quality.

Executive Summary

Anjuna Security sells software that runs unchanged applications inside sealed-off hardware environments from AWS, Intel and AMD, so data stays encrypted while it runs. A multi-party AI clean room and an AI-agent policy layer run on that same base. Its chief executive told Forbes the company did not invent confidential computing and is there to make it extremely easy. The enclave comes from the chipmakers and the clouds, and getting ordinary software into one is Anjuna's work. It says three of the world's top 10 banks use its technology and names none of them. The U.S. Navy and Parfin appear by name, and Anjuna had raised $67 million by August 2024. Anjuna documents a consistent operational model across AWS, Azure and Google Cloud.

Sourced Details

Description Sells software that runs unmodified container applications inside hardware trusted execution environments on AWS, Azure and Google Cloud, and layers AI clean rooms and agent policy enforcement on that runtime. [f1]
Founded 2018 [f2]
HQ Palo Alto, California, United States [f3]
Funding $67M total [f2]
Latest funding Series B2, $25M, August 2024 [f4]

Products

Product What it does
Anjuna Seaglass Runs container and Kubernetes applications inside hardware trusted execution environments on AWS, Azure, Google Cloud and on-premises, with no application code changes.
Anjuna Northstar Clean room built on Seaglass where several parties merge sensitive data and AI models for joint analysis without exposing their own inputs to each other.
Anjuna Overwatch Policy layer that runs inside an enclave, reads every AI agent message and tool call, and brokers the credentials the agent would otherwise hold itself.

Matrix Coverage

AI Defense Matrix

GovernIdentifyProtectDetectRespondRecover
AI-Workload Platforms Inference servers, training platforms, vector DB platforms, and the model-loading supply chain.
AI Orchestration Tools Agentic orchestration tools, plus their plugins, skills, hooks, system prompts, scaffolding, harnesses, configuration settings, and MCP clients on user devices.
AI-Generated Code Code produced by AI tools, AI-assisted reviews, AI-generated infrastructure-as-code and tests, and vibe-coded apps that bypass CI/CD.
AI Gateways & Routers MCP proxies and gateways, LLM routers, outbound AI-service traffic, shadow AI egress, and model-registry traffic.
AI Model Model weights, fine-tuning checkpoints, model cards, registries, AIBOM, and the third-party LLMs your enterprise consumes.
Training Data Datasets used for training, fine-tuning, and continued learning.
Runtime AI Data User prompts, inference inputs, RAG content, vector DB content, persistent agent memory, and interaction history.
AI Agent Identities AI agents as non-human principals, plus credentials, keys, permission scopes, service accounts, and delegation chains across agents and tools.

Anjuna Seaglass runs AI workloads inside trusted execution environments, which give sensitive data and ML models hardware-level isolation, confidentiality and integrity protection. Anjuna Seaglass is mapped to the AI Defense Matrix. [f5]

Cyber Defense Matrix

IdentifyProtectDetectRespondRecover
Devices Workstations, servers, phones, tablets, storage, network devices, IoT infrastructure, and similar hardware.
Applications Software, interactions, and application flows on the devices.
Networks Connections and traffic flowing among devices and apps, plus communication paths.
Data Content at rest, in transit, or in use across devices, apps, and networks.
Users The people using the devices, apps, networks, and data.

Anjuna Seaglass runs container applications inside hardware trusted execution environments, keeping their code and data encrypted while processing and out of reach of privileged administrators. Anjuna Seaglass is mapped to the Cyber Defense Matrix. [f1]

Market Readiness

How well the company can compete in its security market, scored across eight dimensions against public evidence.

Established 25 /40 Established: Market readiness of 25 to 30, the typical band where most analyzed companies land.
Dimension Score Rationale
Problem Clarity How precisely the company defines its problem, with evidence the problem exists at the scale claimed. 3/5 Anjuna names a specific buyer, the organization that wants sensitive work to run on infrastructure it does not own. Both Forbes and Fast Company describe the fragmentation Anjuna sells against, and Forbes names Intel, Microsoft and ARM as the source of the incompatible tooling. No source in the reviewed record quantifies what exposure during processing costs that buyer. [s1, s2, s11, s12]
Capability Depth How specific the technical capabilities are, with evidence beyond marketing claims such as docs and third-party validation. 3/5 Anjuna publishes five documentation sets, covering AWS Nitro, AMD SEV, Kubernetes on SEV, Intel SGX and the Policy Manager, and its AWS Nitro manual documents the architecture and includes a command-line reference. No source code implementing the runtime, no third-party technical evaluation and no customer technical writeup appears in the reviewed record, so vendor documentation alone evidences that depth, without the external validation point a higher score needs. [s15, s16, s2]
Market Timing Whether the market is ready for this product, with evidence that buyers are actively seeking solutions. 3/5 The hardware that makes this possible is dated and independently reported: Forbes wrote in August 2024 that Intel, Microsoft and ARM had already embedded these zones in their CPUs and GPUs while their software stayed incompatible. Anjuna shipped into that gap, taking Northstar to general availability in December 2024 and announcing Overwatch in June 2026, and apart from Fast Company's 2023 note the buyer-side evidence comes from Anjuna's own announcements. [s11, s23, s8, s7]
Team Credibility Demonstrated domain expertise with public signals such as prior exits, publications, and industry recognition. 4/5 Ayal Yogev led the Umbrella product management team at OpenDNS before Cisco bought it for $635 million and was at Imperva through its 2011 IPO, and Mark Bower ran product strategy at Voltage Security before HPE acquired it. Anjuna's team page documents those specific build and exit roles, and Forbes independently corroborates Yogev's enterprise-security employment history, his Unit 8200 service and Yan Michalevsky as his co-founder. [s5, s11, s12]
GTM Proof Evidence of actual traction (customers, revenue signals, partnerships) beyond stated intentions. 3/5 Anjuna publishes a Parfin case study with its co-founder quoted, lists the U.S. Navy, introduced JUMO and Ascendo AI as Northstar design partners, and reaches buyers through an Azure Marketplace listing Microsoft documents. Its larger claims, three of the world's top 10 banks and five payment companies, are counts Anjuna states itself. Outside corroboration of traction in the reviewed record comes down to Fast Company's 2023 note that the company had found customers in banking, defense and blockchain. [s17, s7, s23, s12, s1, s13]
Funding Efficiency Whether funding matches go-to-market ambition, with signs of capital-efficient growth. 3/5 Anjuna had raised $67 million by August 2024, when it last disclosed a round, a $25 million Series B extension proportional to its stage. It kept shipping afterwards, with Northstar generally available that December and Overwatch announced in June 2026. Its August 2024 announcement states that bookings and customer base more than doubled year over year, a vendor-reported figure, and no revenue or margin measure appears in the reviewed record. [s11, s6, s23, s8]
Category Clarity Whether the company creates or fits a recognizable category that buyers can quickly place in their stack. 3/5 Confidential computing is a category outsiders track, with CB Insights publishing a named peer set around Anjuna, and Anjuna announced in January 2025 that Gartner had named it a Tech Innovator in preemptive cybersecurity. Anjuna still sells through vendor-specific labels, a universal confidential computing platform, an AI data fusion clean room and a trusted control plane, so a buyer needs the vendor's explanation to place it. [s22, s9, s1, s8]
Incumbent Defensibility How vulnerable the core value proposition is to absorption as a feature by a platform vendor. 3/5 The enclaves belong to Intel, AMD and the cloud providers, and Microsoft's documentation describes Anjuna as running applications on Azure confidential computing instances powered by AMD SEV-SNP or Intel SGX processors, so what Anjuna sells runs on hardware its suppliers already ship. Its friction is a consistent operational model across AWS, Azure and Google Cloud with attestation-gated secret release on each, which is more than a bundled feature replicates and less than a structural moat. [s13, s11, s1, s2]
Business Risks AWS, Microsoft or Google could extend its own confidential-computing tooling to cover the unchanged-application case Anjuna sells, removing the reason to buy a separate runtime on that cloud…
  • AWS, Microsoft or Google could extend its own confidential-computing tooling to cover the unchanged-application case Anjuna sells, removing the reason to buy a separate runtime on that cloud.
  • Anjuna could reach three years without a disclosed round, having last raised a $25 million Series B extension in August 2024, leaving it to fund three product lines on capital raised before two of them reached market.
  • The three top 10 banks Anjuna counts could stay unnamed, leaving its largest traction claim something a prospect cannot check.
  • Anjuna Overwatch could remain in the early-feedback phase it entered in June 2026, leaving the agent-governance line without a generally available product.
  • Anjuna could go another year without publishing a SOC 2 report or an ISO 27001 certificate, leaving its own control evidence outside the public record while it sells to banks and a defense customer.
Problem & Market Anjuna sells to organizations that want sensitive work to run on infrastructure they do not own…

Anjuna sells to organizations that want sensitive work to run on infrastructure they do not own. Its answer is confidential computing, which Anjuna's own explainer describes as hardware-rooted trusted execution environments that isolate code, data and secrets at runtime, so that neither the host operating system, the hypervisor, nor a cloud provider's administrators can inspect what runs inside.

By August 2024 the obstacle Forbes described was software rather than hardware availability. Intel, Microsoft and ARM had already embedded these zones in their CPUs and GPUs, and the software each of them built to expose the feature was often not compatible with the others. Fast Company described the same fragmentation nine months earlier and framed Anjuna's answer as virtualizing those processors within a single software stack, so an application need not be recoded for each server platform and cloud.

The urgency stays qualitative. No source in the reviewed record puts a number on what exposure during processing costs a bank or a defense agency, so a buyer decides on confidentiality and compliance arguments rather than on a measured loss. [s1, s11, s12, s2]

Product Capabilities Anjuna Seaglass is the platform that carries the rest of the portfolio…

Anjuna Seaglass is the platform that carries the rest of the portfolio. Anjuna says it turns a container application into what it calls a Confidential Container, an enclave-ready hardened image that deploys with one command across AWS, Azure, Google Cloud and on-premises hardware. The Anjuna Policy Manager then releases secrets and configuration to that image only after hardware-based attestation succeeds.

Anjuna's own AWS manual states what the product removes more plainly than its marketing does. An AWS Nitro Enclave gives an application a single socket to another virtual machine instance rather than an internet-connected network interface. Developers must therefore design a protocol to pass data in and out of the enclave and update their applications to use it. The manual calls that a significant engineering effort and says it can even be impossible for closed-source applications. The Anjuna Nitro Runtime runs the application inside the enclave without those changes.

Two products build on that base. Anjuna Northstar, generally available since December 2024, is a clean room where several parties merge data and AI models and no party sees another's raw inputs. Anjuna Overwatch, announced in June 2026, runs and is attested inside an enclave, and it sees every LLM message and tool call an agent makes. It validates the tool calls and data requests against a policy manifest, and it holds the agent's API keys and tokens so the agent never carries them itself.

Anjuna documents the platform in detail: five documentation sets covering AWS Nitro, AMD SEV, Kubernetes on SEV, Intel SGX and the Policy Manager, with the Nitro manual carrying its own command-line reference. What the reviewed record does not contain is an outside assessment of it. No source code implementing the runtime appears there, and neither does a third-party technical evaluation. [s2, s15, s16, s3, s4, s23, s1, s8]

Competitive Positioning CB Insights lists nine top competitors for Anjuna…

CB Insights lists nine top competitors for Anjuna. Its named comparisons include Opaque, Cysec, Repello AI, Phala, Enveil, Edgeless Systems and Fortanix.

Anjuna's own suppliers pose the harder competitive question. Microsoft's own documentation describes Anjuna as software that lets you run applications on Azure confidential computing instances powered by AMD SEV-SNP or Intel SGX processors. That is an accurate description of a supplier relationship, and of an exposure: the party selling the enclave can also sell the tooling around it. Anjuna's chief executive stated the boundary himself, telling Forbes the company did not invent confidential computing and is there to make it extremely easy.

Sameness across clouds is what Anjuna offers against that. Anjuna documents one operational model across AWS, Azure and Google Cloud, with attestation-gated secret release on each, which is what a buyer gives up by taking a single provider's native tooling instead. [s22, s13, s11, s1, s2]

Go-to-Market & Traction Anjuna's largest traction claims arrive as counts…

Anjuna's largest traction claims arrive as counts. A June 2025 announcement says a new top five bank brought its total to three of the world's top 10 banks, that it also serves five payment companies and the U.S. Navy, and that the headcount of its customers now exceeds one million. It names none of the banks, and its two bank case studies are titled for an international bank and a global bank.

The named references are narrower than the counts imply. Parfin, a digital-asset custody firm, appears with its co-founder quoted about protecting distributed key shares inside secure enclaves. The U.S. Navy appears in a case study and in the August 2024 funding announcement, which describes it testing Llama 3 models on confidential NVIDIA H100 GPUs through NVIDIA LaunchPad. JUMO and Ascendo AI arrived in December 2024 as Northstar design partners, with the announcement describing JUMO using Northstar in its credit-risk workflow.

Outside corroboration is old and general. Fast Company wrote in November 2023 that the approach had helped Anjuna find customers in banking, defense and blockchain, and nothing more recent in the reviewed record confirms traction from a source other than Anjuna.

Anjuna publishes partner pages for AWS Nitro Enclaves and for Microsoft Azure confidential computing, and Microsoft's own partner page routes buyers to an Azure Marketplace SaaS offering and a managed application. The free trial is not self-service: a prospect submits a form and Anjuna responds within two business days with access, and the page asks for comfort working from the shell and with Docker containers. [s7, s17, s1, s6, s23, s12, s18, s13]

Team & Credibility The founders' backgrounds are the most independently corroborated asset in the record…

The founders' backgrounds are the most independently corroborated asset in the record. Forbes reports that Ayal Yogev served in Israel's Unit 8200, studied electrical engineering and computer science at Tel Aviv University, took an MBA at Berkeley, and founded Anjuna in 2018 with Yan Michalevsky while Michalevsky was finishing a doctorate at Stanford. Fast Company separately describes the company as founded by Unit 8200 veterans.

The commercial record behind that is in the same domain Anjuna now sells into. Anjuna's team page records that Yogev led the Umbrella product management team at OpenDNS, which Cisco acquired in 2015 for $635 million, was part of Imperva's 2011 IPO, and was most recently VP of product management at SafeBreach. Mark Bower headed product and business strategy for Voltage Security, acquired by HPE, and the Atalla hardware-security-module business there. Steve Van Lare was VP of engineering at Automation Anywhere and at the fraud-detection company 41st Parameter.

The board and advisors add recognized cryptography and national-security experience. Anjuna's team page carries Stanford cryptographer Dan Boneh, who heads the university's applied cryptography group, and the August 2024 funding announcement records that SineWave's Patricia Muoio, a former senior executive at the National Security Agency, joined the board. [s11, s12, s5, s6]

Trust Readiness No security attestation of Anjuna's own appears in the reviewed public record…

No security attestation of Anjuna's own appears in the reviewed public record. No SOC 2 report, ISO 27001 certificate, FedRAMP authorization or third-party trust portal appears on the probed surfaces or anywhere in the sources reviewed.

A bank or a defense customer that wants Anjuna's own control evidence will not find it in the reviewed sources.

The compliance language Anjuna does publish belongs to the product rather than to the company. Anjuna Overwatch maps agent actions to frameworks such as NIST 800-53 and PCI DSS for reporting to security teams, which serves a customer's compliance posture and says nothing about Anjuna's own.

The adversarial record is quiet. The National Vulnerability Database returns no CVE naming Anjuna, and no breach, enforcement action or critical technical write-up about the company appears in the reviewed sources. [s21, s20, s4, s7]

Competitors Opaque Systems, Fortanix, Phala, Enveil, Edgeless Systems…
Company Relationship Note Compare
Opaque Systems competes with CB Insights names Opaque first among Anjuna's top competitors. N/AWe captured the evidence for these companies under different evidence-model versions (v1 vs v2), so the totals were scored under different conditions and are not directly comparable.
Fortanix competes with Appears in the Anjuna comparison view CB Insights publishes. N/AWe scored these companies at different scopes, so the totals measure different things.
Phala competes with Listed among Anjuna's competitors by CB Insights.
Enveil competes with Listed among Anjuna's competitors by CB Insights.
Edgeless Systems competes with Listed among Anjuna's competitors by CB Insights.

Add analyzed competitors to compare them side by side with Anjuna Security.

Strategy Deep Dive

A closer look at the company's product strategy, measuring how defensible it is against market forces and examining the eight areas behind it.

Defensibility

Contested 14 /21 Contested: Defensibility of 13 to 14, the typical band, where a moat exists but is under pressure. reinforce or reposition

The hardware that makes Anjuna's product work belongs to someone else. Intel, AMD and the cloud providers build the sealed-off environments where data stays encrypted while it runs. Anjuna's chief executive told Forbes the company did not invent confidential computing and is there to make it extremely easy. What Anjuna adds is software that carries an unchanged application into any of those enclaves. Its own AWS manual describes what that removes, a data-passing protocol a developer would otherwise write by hand. That software is demanding systems work a funded rival could still reproduce. The reviewed record documents no dataset Anjuna accumulates and no security certification of its own, so Anjuna has a head start rather than a durable lead.

Dimension Score Rationale
Value Delivery Does the product sell software as the product, or judgment, trust, or accountability with software as the delivery mechanism. 1/3 What Anjuna sells is software: a licensed runtime, a policy manager, a clean-room product and an agent control layer, sold and delivered as product rather than as judgment or retained accountability. On the deployment Anjuna documents, the customer runs the software itself, the software-product level.
Switching Cost How expensive leaving is for a customer: data portability, integrations, learned workflows, network effects, regulatory data residency. 2/3 Leaving Anjuna means rebuilding enclave images, attestation policies and secret-release wiring on a replacement, which is meaningful integration friction. The cited record does not size the migration: it states no exit duration, no multi-party transfer procedure and no artifact that cannot be exported. The engineering effort Anjuna's manual describes is the effort of adopting a raw enclave rather than of leaving Anjuna.
Compliance Moat Whether certifications, liability acceptance, or audit trails block an easy replacement. 1/3 No certification, authorization or liability acceptance of Anjuna's own appears on the probed surfaces or in the reviewed sources. The compliance mapping Anjuna publishes belongs to the product, with Overwatch reporting agent actions against frameworks such as NIST 800-53 and PCI DSS, which serves the customer's obligations and blocks no replacement.
Problem Complexity Whether the product requires ML, optimization, real-time systems, or years of specialized expertise. 3/3 Software that carries unchanged applications into AWS Nitro, AMD SEV, Intel SGX and Kubernetes, with attestation-gated secret release across all of them, is demanding systems engineering. Forbes described the work as sitting deep in the system stack, and Anjuna's chief executive told it the company is almost an operating-system company. Five manuals span four deployment environments plus the Policy Manager, and the Nitro manual carries its own command reference.
Buyer Profile Whether buyers are SMB operators, mid-market IT teams, or regulated enterprises and governments with procurement gates. 3/3 The evidenced buyers are regulated enterprises and government. Anjuna states that three of the world's top 10 banks and five payment companies use its technology, names the U.S. Navy in a case study and in its funding announcement, and Fast Company independently placed its customers in banking, defense and blockchain. The banks are counted rather than named, which limits what a prospect can verify without changing the buyer class.
Layer Whether the product is an end-user application, a platform with application features, or infrastructure other applications depend on. 3/3 The customer's application executes inside the environment Seaglass creates, so removing Anjuna stops the confidential workload rather than removing a control beside it. Northstar and Overwatch both run on Seaglass, which puts it at the infrastructure level other software depends on.
Proprietary Data, Content, or IP Whether the product accumulates datasets, content licenses, or IP that a rival cannot recreate from scratch. 1/3 No named non-public dataset, content license or patent of Anjuna's own appears in the reviewed record, and the reviewed record identifies no cross-customer dataset or other accumulated data asset. The differentiating asset is the runtime engineering, which a funded rival could reproduce against the same publicly documented enclave interfaces.
Strategic Market Segmentation Anjuna sells to organizations whose sensitive work has to run on infrastructure they do not control, and the published examples include financial services, government, healthcare, SaaS and blockchain…

Anjuna sells to organizations whose sensitive work has to run on infrastructure they do not control, and the published examples include financial services, government, healthcare, SaaS and blockchain. The August 2024 funding announcement puts it plainly: Anjuna works with enterprises in financial services, government and blockchain.

Financial services is where the record is thickest. A June 2025 announcement claims three of the world's top 10 banks, five payment companies and a combined customer headcount above one million. The public evidence for government adoption is concentrated in the single U.S. Navy reference.

The segmentation shifted with the products. Seaglass addresses an infrastructure team moving existing applications. Northstar addresses data scientists and AI operations teams who need several parties to pool data. Overwatch addresses a security team that has to govern autonomous agents. Those are three different technical audiences for one platform.

Product Capabilities & AI Advantages Anjuna says Seaglass turns a container application into an enclave-ready hardened image and deploys it with one command across AWS, Azure, Google Cloud, on-premises hardware and Kubernetes…

Anjuna says Seaglass turns a container application into an enclave-ready hardened image and deploys it with one command across AWS, Azure, Google Cloud, on-premises hardware and Kubernetes. The Anjuna Policy Manager then releases secrets to that image only after hardware-based attestation succeeds. The vendor's own claim for the result is a 90 percent reduction in time to success on container applications.

What the platform is doing is best read from Anjuna's AWS manual rather than its marketing. An AWS Nitro Enclave gives an application a single socket to another virtual machine instance rather than an internet-connected network interface. Developers must therefore design a protocol to pass data in and out and update their applications to use it. The manual calls that a significant engineering effort and says it can even be impossible for closed-source applications. The Anjuna Nitro Runtime removes that work by running the application inside the enclave unchanged. Microsoft's own partner documentation describes the same relationship on Azure: Anjuna runs applications on Azure confidential computing instances powered by AMD SEV-SNP or Intel SGX processors.

The two AI products build on that runtime. Anjuna Northstar, generally available since December 2024, creates clean rooms where several parties merge datasets and AI models, work with them in ordinary Jupyter notebooks, and see only their own inputs. Anjuna Overwatch, announced in June 2026, runs attested inside an enclave and sees every model message and tool call an agent makes. It validates the tool calls and data requests against a policy manifest that covers MCP server access. It brokers the agent's API keys and tokens so the agent never holds them, and it redacts personal data and secrets before they reach models, logs or memory stores.

Anjuna documents the platform in detail. It publishes five documentation sets, covering AWS Nitro, AMD SEV, Kubernetes on SEV, Intel SGX and the Policy Manager, and its AWS Nitro manual documents the architecture and includes a command-line reference. No source code implementing the runtime appears in the reviewed sources, and no third-party technical evaluation of it does either, so every quantitative performance claim traces back to Anjuna itself. Microsoft's partner page, SecurityWeek and Fast Company all describe the software from outside, but none of them is a technical evaluation of it.

Sales Engagement & Go-to-Market The motion is enterprise sales with a gated trial at the front…

The motion is enterprise sales with a gated trial at the front. The free-trial page offers all Seaglass features for 30 days, but a prospect submits a form and Anjuna answers within two business days with access, and the page asks for comfort working from the shell. That is approval-gated rather than self-service, and the same page asks for comfort working with Docker containers, so the trial is aimed at a platform engineer rather than at a casual evaluator.

Anjuna's published partnerships are with the companies that supply the hardware it runs on. Anjuna publishes partner pages for AWS Nitro Enclaves and for Microsoft Azure confidential computing, Microsoft's own partner page routes buyers to an Azure Marketplace listing, and the NVIDIA relationship is the one with a named outcome behind it: the U.S. Navy tested Llama 3 models on confidential NVIDIA H100 GPUs using Seaglass through NVIDIA LaunchPad.

Proof arrives mostly as counts Anjuna states itself. Three of the world's top 10 banks, five payment companies, a customer headcount above one million, and bookings and customer base more than doubling year over year in the 2024 announcement are all first-party. The named references are Parfin, quoted on protecting distributed key shares, and the U.S. Navy. JUMO and Ascendo AI arrived as Northstar design partners, with the December 2024 announcement describing JUMO using Northstar in its credit-risk workflow. Fast Company's 2023 note that the approach had found customers in banking, defense and blockchain is the outside corroboration on the record, and nothing more recent joins it.

Pricing Model Nothing in the reviewed sources states a list rate, a packaging tier or a charging unit for any of the three products…

Nothing in the reviewed sources states a list rate, a packaging tier or a charging unit for any of the three products. On Anjuna's own site a prospect can request a demo, contact sales or apply for the gated 30-day trial, and Microsoft's partner page adds two more routes, an Azure Marketplace SaaS offering and a managed application.

The trial is scoped to one cloud platform the prospect picks from AWS with Nitro Enclaves, Azure with AMD SEV or Google Cloud with AMD SEV. Anjuna's documentation includes a licensing step for the Nitro runtime, and the reviewed sources say nothing about how a production license is counted or enforced.

Two consequences follow for a buyer. No rate appears in the reviewed sources, so a prospect cannot size a deal from the reviewed sources. And the vendor's published economics claim, that customers saw a 12 percent reduction in security spending, is a return argument rather than a price, so it cannot be checked against a rate card.

Product Delivery & Operations The customer runs the deployment Anjuna documents…

The customer runs the deployment Anjuna documents. Seaglass builds an enclave-ready image from the customer's own container, deploys it with one command into the customer's cloud account or data center, and runs there. The Policy Manager releases secrets to it only once hardware-based attestation succeeds. Microsoft's partner page also routes buyers to an Azure Marketplace SaaS offering and a managed application, and the reviewed record does not say who operates those.

That places the operational burden on the buyer's platform team, and Anjuna's documentation reflects it: documentation sets for AWS Nitro, AMD SEV, Kubernetes on SEV and Intel SGX, Kubernetes pod specifications, performance and resource guidelines, a troubleshooting guide and a command reference. The trial page asks for exactly the skills those manuals assume.

Portability is the operational selling point. One operational model across AWS, Azure, Google Cloud and on-premises, and the ability to run Overwatch from cloud to air-gapped environments, means a platform team learns the tooling once rather than once per provider.

Earning Customers' Trust No security attestation of Anjuna's own appears in the reviewed public record…

No security attestation of Anjuna's own appears in the reviewed public record. No SOC 2 report, ISO 27001 certificate, FedRAMP authorization or third-party trust portal appears on the probed surfaces or anywhere in the sources reviewed, which is a gap against the buyers the company names.

The compliance material Anjuna does publish belongs to the product. Overwatch maps agent actions to frameworks such as NIST 800-53 and PCI DSS and reports them to security teams. Anjuna's NVIDIA partner page says the joint solution lets enterprises validate the integrity and trustworthiness of large language models through remote attestation, simplifying regulatory compliance. Both serve a customer's obligations. Neither is a statement about Anjuna's own controls.

The product's assurance is cryptographic. Attestation is the mechanism: the Policy Manager releases secrets to a workload only after hardware-based proof that the expected code is running in a genuine enclave. That answers nothing about how Anjuna runs its own business.

The adversarial record is quiet. The National Vulnerability Database returns no CVE naming Anjuna, and no breach, enforcement action or critical technical write-up about the company appears in the reviewed sources.

Platform Strategy & Ecosystem Positioning Anjuna is building a platform in the literal sense: Seaglass is the base, and both Northstar and Overwatch run on it…

Anjuna is building a platform in the literal sense: Seaglass is the base, and both Northstar and Overwatch run on it. Overwatch is attested inside a trusted execution environment powered by Seaglass, and Northstar is described as powered by Seaglass, so all three products build on one engineering effort.

Its ecosystem position is dependent by construction. The enclaves come from Intel, AMD and the cloud providers, and Anjuna's own partner pages cover those same suppliers plus NVIDIA, whose Blackwell and Hopper GPUs it pairs with Seaglass for confidential AI. The enclave technology comes from those suppliers. Microsoft documents its own confidential-computing instances and routes buyers to a marketplace listing, so Anjuna's contribution is sameness across clouds.

The agentic-AI line is the newest bet on the platform. Anjuna's solution page divides the work explicitly, with Seaglass providing hardware-rooted identity, runtime attestation and encrypted execution for agents and MCP servers, Northstar governing MCP access and isolating shared data, and Overwatch supervising behavior at runtime. Overwatch entered an early-feedback phase in June 2026, and the reviewed record names no production customer for it.

Team & Execution Capability The founding story is corroborated from outside…

The founding story is corroborated from outside. Forbes reports that Ayal Yogev served in Israel's Unit 8200, studied electrical engineering and computer science at Tel Aviv University, took an MBA at Berkeley, and founded Anjuna in 2018 with Yan Michalevsky while Michalevsky was completing a doctorate at Stanford. Fast Company separately describes the company as founded by Unit 8200 veterans.

The executive record is in the domain Anjuna sells into. Yogev led the Umbrella product management team at OpenDNS, which Cisco acquired in 2015 for $635 million, was part of Imperva's 2011 IPO, and was VP of product management at SafeBreach. Mark Bower headed product and business strategy for Voltage Security, acquired by HPE, and the Atalla hardware-security-module business there, and most recently led product management for Amazon's Payment Cryptography business. Steve Van Lare was VP of engineering at Automation Anywhere and at the fraud-detection company 41st Parameter.

The advisory roster carries recognized academic cryptography. Anjuna's team page carries Stanford cryptographer Dan Boneh, who heads the university's applied cryptography group, and Chia-Che Tsai, the main developer of the Graphene library operating system for isolating Linux applications on Intel SGX. The August 2024 announcement records that SineWave's Patricia Muoio, formerly a senior executive at the National Security Agency, joined the board alongside M Ventures' Ilja Aizenberg.

Sources

Company Detail Sources (5)
Id Source Tier Accessed
f1 Anjuna Seaglass: Universal Confidential Computing official 2026-08-17
f2 Forbes: This Startup Raised $25 Million To Keep Data Safe While Training AI press 2026-08-17
f3 SEC EDGAR filing record for Anjuna Security, Inc., CIK 0001870313 regulatory 2026-08-17
f4 Anjuna Raises $25M Funding to Fuel Momentum and Expand Confidential Computing Offerings for AI official 2026-08-17
f5 Anjuna Secure AI solution page official 2026-08-17
Profile Analysis Sources (24)
Id Source Tier Accessed
s1 Anjuna home page: product lineup, confidential-computing FAQ, and customer quotes
“Confidential Computing uses hardware-rooted Trusted Execution Environments (TEEs), also called secure enclaves, to isolate code, data, and secrets at runtime.”
official 2026-08-17
s2 Anjuna Seaglass product page: containers in trusted execution environments, attestation, and multi-cloud reach
“Anjuna Seaglass delivers frictionless confidential computing for containers and Kubernetes applications across any cloud or on-premises infrastructure, deployable in minutes.”
official 2026-08-17
s3 Anjuna Northstar product page: multi-party AI data clean rooms
“Anjuna Northstar enables interactive multi-party data collaboration and AI-driven dataset analyses to discover outcomes without compromising data privacy or intellectual property.”
official 2026-08-17
s4 Anjuna Overwatch product page: runtime policy enforcement for AI agents
“Anjuna Overwatch delivers an intelligent, confidential computing-protected supervisory layer that guards and governs agent behavior, data use, compliance, and cost.”
official 2026-08-17
s5 Anjuna leadership, board, and advisor page
“Ayal has 20 years of experience in the enterprise security market, serving most recently as VP of Product Management at SafeBreach. Prior to that, he led the Umbrella product management team at OpenDNS, which was acquired by Cisco in 2015 for $635M.”
official 2026-08-17
s6 Anjuna announcement of its August 2024 Series B2 round
“today announced that it has closed a $25 million Series B2 financing round to propel and expand its AI-enabling enterprise offerings.”
official 2026-08-17
s7 Anjuna announcement of a third top-ten global bank, June 2025
“This win brings Anjuna’s total to three of the world’s top 10 banks using its technology to securely collaborate and unlock data value with confidence.”
official 2026-08-17
s8 Anjuna announcement introducing Anjuna Overwatch, June 2026
“today announced the availability of an industry-unique solution for managing risk, controlling usage, and securing agentic AI.”
official 2026-08-17
s9 Anjuna announcement of its listing as a Tech Innovator in a Gartner report
“announced today that it has been named a Tech Innovator in Preemptive Cybersecurity in the 2024 Gartner Emerging Tech: Tech Innovators in Preemptive Cybersecurity report .”
official 2026-08-17
s10 SecurityWeek report by Ionut Arghire on Anjuna's June 2021 Series B
“Anjuna, a provider of cloud application security, today announced that it has raised $30 million in Series B funding, which brings the total raised by the company to date to $42 million.”
press 2026-08-17
s11 Forbes report by Kirk Ogunrinde on Anjuna's August 2024 round
“"We've essentially built what we call a universal confidential computing platform," cofounder and CEO Ayal Yogev told Forbes. "We didn't invent confidential computing — we're just here to make it extremely easy."”
press 2026-08-17
s12 Fast Company write-up by Alex Pasternack naming Anjuna a Next Big Things in Tech honoree
“Founded by veterans of Israel’s Unit 8200, Anjuna attempts to tear through the mess by virtualizing confidential computing CPUs within a single software stack, thus avoiding the need to recode individual applications for every server infrastructure and cloud service.”
press 2026-08-17
s13 Microsoft Learn Azure confidential-computing partner page for Anjuna
“It allows you to run applications on Azure confidential computing instances powered by AMD SEV-SNP or by Intel SGX CPUs with enhanced ease of use, operational efficiency and security posture.”
other 2026-08-17
s14 SEC EDGAR filing record for Anjuna Security, Inc., CIK 0001870313
“"name":"Anjuna Security, Inc."”
regulatory 2026-08-17
s15 Anjuna Nitro Runtime documentation, introduction and command reference
“allows you to run code in a secure enclave with no external access to the enclave’s memory and CPU.”
official 2026-08-17
s16 Anjuna documentation index listing the per-platform runtime manuals
“AWS Nitro documentation”
official 2026-08-17
s17 Anjuna customers page listing its published case studies
“U.S. Navy Charters Secure AI Course with Confidential Computing”
official 2026-08-17
s18 Anjuna free-trial page: what the trial covers and how a request is handled
“Access to all Anjuna Seaglass features for 30 days”
official 2026-08-17
s19 Anjuna agentic-AI solution page describing how its three products divide the work
“Anjuna's agentic AI security makes your operations safer by isolating agents, data, and MCP servers in Trusted Execution Environments (TEEs).”
official 2026-08-17
s20 NVD keyword probe for Anjuna on 2026-08-17, zero records returned (a Fortanix control query on the same endpoint returned three)
“"resultsPerPage":0,"startIndex":0,"totalResults":0,"format":"NVD_CVE"”
other 2026-08-17
s24 Anjuna NVIDIA partner page: confidential AI on Blackwell and Hopper GPUs
“With the revolutionary Blackwell and Hopper architecture GPUs, NVIDIA unlocked a new era of safeguarding "data-in-use" during GPU processing. Paired with Anjuna Seaglass, our partnership guarantees data confidentiality and integrity throughout the entire AI workload lifecycle.”
official 2026-08-17
s23 Anjuna announcement that Anjuna Northstar reached general availability, December 2024
“PALO ALTO, Calif., Dec 2, 2024 – Anjuna , creator of Anjuna Seaglass, the Universal Confidential Computing Platform, today announced the general availability of Anjuna Northstar , the world’s first AI Data Fusion Clean Room.”
official 2026-08-17
s22 CB Insights alternatives-and-competitors listing for Anjuna
“Anjuna 's top competitors include Opaque , Cysec , and Repello AI .”
research 2026-08-17
s21 Trust-surface probe 2026-08-17: trust. and security. subdomains do not resolve, nor does a nonsense control, and /trust, /security, /compliance 404 official 2026-08-17
Deep-Dive Sources (24)
Id Source Tier Accessed
s1 Anjuna home page: product lineup, confidential-computing FAQ, and customer quotes
“Confidential Computing uses hardware-rooted Trusted Execution Environments (TEEs), also called secure enclaves, to isolate code, data, and secrets at runtime.”
official 2026-08-17
s2 Anjuna Seaglass product page: containers in trusted execution environments, attestation, and multi-cloud reach
“Anjuna Seaglass delivers frictionless confidential computing for containers and Kubernetes applications across any cloud or on-premises infrastructure, deployable in minutes.”
official 2026-08-17
s3 Anjuna Northstar product page: multi-party AI data clean rooms
“Anjuna Northstar enables interactive multi-party data collaboration and AI-driven dataset analyses to discover outcomes without compromising data privacy or intellectual property.”
official 2026-08-17
s4 Anjuna Overwatch product page: runtime policy enforcement for AI agents
“Anjuna Overwatch delivers an intelligent, confidential computing-protected supervisory layer that guards and governs agent behavior, data use, compliance, and cost.”
official 2026-08-17
s5 Anjuna leadership, board, and advisor page
“Ayal has 20 years of experience in the enterprise security market, serving most recently as VP of Product Management at SafeBreach. Prior to that, he led the Umbrella product management team at OpenDNS, which was acquired by Cisco in 2015 for $635M.”
official 2026-08-17
s6 Anjuna announcement of its August 2024 Series B2 round
“today announced that it has closed a $25 million Series B2 financing round to propel and expand its AI-enabling enterprise offerings.”
official 2026-08-17
s7 Anjuna announcement of a third top-ten global bank, June 2025
“This win brings Anjuna’s total to three of the world’s top 10 banks using its technology to securely collaborate and unlock data value with confidence.”
official 2026-08-17
s8 Anjuna announcement introducing Anjuna Overwatch, June 2026
“today announced the availability of an industry-unique solution for managing risk, controlling usage, and securing agentic AI.”
official 2026-08-17
s9 Anjuna announcement of its listing as a Tech Innovator in a Gartner report
“announced today that it has been named a Tech Innovator in Preemptive Cybersecurity in the 2024 Gartner Emerging Tech: Tech Innovators in Preemptive Cybersecurity report .”
official 2026-08-17
s10 SecurityWeek report by Ionut Arghire on Anjuna's June 2021 Series B
“Anjuna, a provider of cloud application security, today announced that it has raised $30 million in Series B funding, which brings the total raised by the company to date to $42 million.”
press 2026-08-17
s11 Forbes report by Kirk Ogunrinde on Anjuna's August 2024 round
“"We've essentially built what we call a universal confidential computing platform," cofounder and CEO Ayal Yogev told Forbes. "We didn't invent confidential computing — we're just here to make it extremely easy."”
press 2026-08-17
s12 Fast Company write-up by Alex Pasternack naming Anjuna a Next Big Things in Tech honoree
“Founded by veterans of Israel’s Unit 8200, Anjuna attempts to tear through the mess by virtualizing confidential computing CPUs within a single software stack, thus avoiding the need to recode individual applications for every server infrastructure and cloud service.”
press 2026-08-17
s13 Microsoft Learn Azure confidential-computing partner page for Anjuna
“It allows you to run applications on Azure confidential computing instances powered by AMD SEV-SNP or by Intel SGX CPUs with enhanced ease of use, operational efficiency and security posture.”
other 2026-08-17
s14 SEC EDGAR filing record for Anjuna Security, Inc., CIK 0001870313
“"name":"Anjuna Security, Inc."”
regulatory 2026-08-17
s15 Anjuna Nitro Runtime documentation, introduction and command reference
“allows you to run code in a secure enclave with no external access to the enclave’s memory and CPU.”
official 2026-08-17
s16 Anjuna documentation index listing the per-platform runtime manuals
“AWS Nitro documentation”
official 2026-08-17
s17 Anjuna customers page listing its published case studies
“U.S. Navy Charters Secure AI Course with Confidential Computing”
official 2026-08-17
s18 Anjuna free-trial page: what the trial covers and how a request is handled
“Access to all Anjuna Seaglass features for 30 days”
official 2026-08-17
s19 Anjuna agentic-AI solution page describing how its three products divide the work
“Anjuna's agentic AI security makes your operations safer by isolating agents, data, and MCP servers in Trusted Execution Environments (TEEs).”
official 2026-08-17
s20 NVD keyword probe for Anjuna on 2026-08-17, zero records returned (a Fortanix control query on the same endpoint returned three)
“"resultsPerPage":0,"startIndex":0,"totalResults":0,"format":"NVD_CVE"”
other 2026-08-17
s24 Anjuna NVIDIA partner page: confidential AI on Blackwell and Hopper GPUs
“With the revolutionary Blackwell and Hopper architecture GPUs, NVIDIA unlocked a new era of safeguarding "data-in-use" during GPU processing. Paired with Anjuna Seaglass, our partnership guarantees data confidentiality and integrity throughout the entire AI workload lifecycle.”
official 2026-08-17
s23 Anjuna announcement that Anjuna Northstar reached general availability, December 2024
“PALO ALTO, Calif., Dec 2, 2024 – Anjuna , creator of Anjuna Seaglass, the Universal Confidential Computing Platform, today announced the general availability of Anjuna Northstar , the world’s first AI Data Fusion Clean Room.”
official 2026-08-17
s22 CB Insights alternatives-and-competitors listing for Anjuna
“Anjuna 's top competitors include Opaque , Cysec , and Repello AI .”
research 2026-08-17
s21 Trust-surface probe 2026-08-17: trust. and security. subdomains do not resolve, nor does a nonsense control, and /trust, /security, /compliance 404 official 2026-08-17

Disclaimer

This content is provided "as is" with no warranties.

This site is an experimental research aid created by Zeltser Security Corp. All its data gathering and analysis was performed autonomously without human review, and it can contain errors of fact, interpretation, and judgment that a human reviewer might catch.

The analyses are statements of opinion, not statements of fact. Machine analysis produced the scores, summaries, and matrix placements by weighing the public sources each page cites, and reasonable people can weigh the same sources differently. Where a page states a fact, it cites the public source and the date it was checked, and the statement is only as accurate as that source. Unless a profile expressly says otherwise, the analysis involves no hands-on testing and no independent validation of any company's products or services.

Nothing here is professional, security, legal, financial, investment, or purchasing advice, and nothing here is a recommendation to invest in, do business with, or avoid any company. Inclusion of a company is not an endorsement, and absence of a company is not a judgment about it. Reading this site creates no advisory or client relationship. Verify any detail you plan to act on against the vendor's current materials.

The content is provided "as is" and "as available," with all warranties disclaimed, express or implied, including merchantability, fitness for a particular purpose, accuracy, and non-infringement. No entry is warranted to be complete, current, or correct. Companies change, vendors update their claims, sources can be wrong, and automated analysis can misread them.

To the fullest extent permitted by law, the operator, Zeltser Security Corp, is not liable for any damages that arise from using this site or relying on its content, including direct, indirect, incidental, special, and consequential damages and lost profits, even if advised that such damages were possible. If you are dissatisfied with the site or disagree with these terms, your remedy is to stop using it.

Entries link to vendor pages, press coverage, and other external sites that Zeltser Security Corp does not control and is not responsible for. A link is not an affiliation with the destination or an endorsement of it. Product and company names and trademarks are the property of their owners, used here nominatively to identify the companies described. Short quotations from cited sources appear for identification and commentary.

Use, quotation, automated retrieval, and redistribution of the content are governed by the Terms of Use at cybercompanyprofiles.com/terms, which permit personal and internal business use with attribution and prohibit republication and resale.