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.
Tenet Security sells security teams a platform that finds the AI agents on company endpoints and in the cloud and blocks their harmful actions before they execute. Tenet says its patent-pending simulation predicts each agent’s next move. Tenet’s founders built Cisco’s AI Defense research team and earlier ran Wild Pointer, a cybersecurity company with Fortune 500 customers. The Westly Group led Tenet’s $6 million seed round, and MizMaa Ventures also invested. Tenet publishes an endorsement from Elevate Textiles’ CISO, whom it also lists as an advisor. It says a legal-sector customer grew from two AI agents to over 20. CrowdStrike says it now secures AI agents while they run, so Tenet’s customers could get overlapping protection from endpoint tools they already use.
| Description | Tenet Security sells a runtime security platform for AI agents that discovers the agents running across endpoints and cloud, reconstructs their sessions, and stops harmful agent actions before they execute. | [f1] |
|---|---|---|
| HQ | New York, United States | [f2] |
| Funding | $6M total | [f3] |
| Latest funding | Seed, $6M led by The Westly Group and MizMaa Venture (June 2026) | [f3] |
| Product | What it does |
|---|---|
| Tenet Security Platform | Runtime security platform that discovers AI agents and their MCP and tool connections, reconstructs agent sessions, blocks harmful agent actions, and red-teams agents before production. |
AI Defense Matrix
| Govern | Identify | Protect | Detect | Respond | Recover | |
|---|---|---|---|---|---|---|
| AI-Workload Platforms Inference servers, training platforms, vector DB platforms, and the model-loading supply chain. | ||||||
| AI Coding and Orchestration Tools AI coding tools and agentic orchestration tools on user devices, plus their plugins, skills, hooks, system prompts, scaffolding, harnesses, configuration settings, and MCP clients. | ||||||
| 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 Traffic Paths to external and self-hosted AI services, MCP tool channels, agent-to-agent calls, model-registry downloads, and egress to unapproved AI services. AI gateways, LLM routers, and MCP gateways steer traffic along those paths. The steering decisions and their inputs belong here too: routing policies, MCP server registries, and agent naming and discovery services. | ||||||
| AI Model Model weights, fine-tuning checkpoints, model cards, registries, an AI bill of materials (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. |
Tenet Security Platform runs inside the AI agent runtime, simulates where each agent action leads before it executes, and prevents the dangerous ones. These capabilities are mapped to the AI Defense Matrix. [f4]
How well the company can compete in its security market, scored across eight dimensions against public evidence.
| Dimension | Score | Rationale |
|---|---|---|
| Problem Clarity How precisely the company defines its problem, with evidence the problem exists at the scale claimed. | 4/5 | Tenet names its buyer, enterprise security teams deploying agents, and the pain of agents hijacked through data they read. VentureBeat quantifies the gap from 2026 surveys, including an Okta and Apprize360 finding that 34% of organizations control agents as they do humans. A Cloud Security Alliance note adds that most organizations lack policy on which data sources agents may query. [s3, s26, s20, s24] |
| Capability Depth How specific the technical capabilities are, with evidence beyond marketing claims such as docs and third-party validation. | 3/5 | Tenet’s product, how-it-works and FAQ pages describe the sensor, session reconstruction, gateway plugin, adapters and pre-execution simulation in concrete terms. Press coverage relays the company’s own description. The documentation site requires a sign-in, the public GitHub repositories hold hardening and scanning tools rather than the platform, and no third-party test appears in the reviewed sources. [s2, s5, s6, s12, s13, s34, s35] |
| Market Timing Whether the market is ready for this product, with evidence that buyers are actively seeking solutions. | 3/5 | A June 2026 Cloud Security Alliance note says MCP connections by mid-2026 span error trackers, version control and cloud consoles. Buyer-side demand in the reviewed sources is one kind of signal, the 2026 surveys VentureBeat cites, plus press interest in Tenet’s research. [s24, s20, s14, s16] |
| Team Credibility Demonstrated domain expertise with public signals such as prior exits, publications, and industry recognition. | 4/5 | SecurityWeek reports the founders built Cisco’s AI Defense research team. SiliconANGLE and Ctech report they earlier founded Wild Pointer, a cybersecurity company with Fortune 500 customers. Independent outlets thus report two in-domain builds, and SiliconANGLE adds that both founders have spoken at DEF CON and Black Hat. [s12, s13, s22, s21] |
| GTM Proof Evidence of actual traction (customers, revenue signals, partnerships) beyond stated intentions. | 3/5 | Techspective attributes Tenet’s deployment figures to Tenet, and Ynetnews notes the company disclosed no customer names. Tenet names Elevate Textiles under its trusted-by heading. It announced its Claude Partner Network membership in August 2026. [s13, s25, s32, s3, s27] |
| Funding Efficiency Whether funding matches go-to-market ambition, with signs of capital-efficient growth. | 3/5 | Tenet raised a $6 million seed round in June 2026. It ships a platform spanning discovery, detection, blocking and testing, and added Claude coverage in August. No revenue, margin or growth figure appears in the reviewed sources, so output per dollar is unconfirmed. [s22, s12, s2, s27] |
| Category Clarity Whether the company creates or fits a recognizable category that buyers can quickly place in their stack. | 3/5 | Independent outlets place Tenet in AI agent security. VentureBeat says it sells runtime protection for AI agents, and Ctech says it focuses on runtime protection for autonomous systems. The category is still forming, with Techspective noting many funded startups on the problem, and Tenet explains its place against gateways and EDR. [s21, s22, s25, s4] |
| Incumbent Defensibility How vulnerable the core value proposition is to absorption as a feature by a platform vendor. | 3/5 | Tenet covers agents from many vendors, including Cursor, Codex, GitHub Copilot, Bedrock and Azure Foundry. That cross-vendor coverage adds friction to absorption by any single agent vendor. No structural moat appears in the reviewed sources, and VentureBeat quotes CrowdStrike’s chief technology officer saying CrowdStrike now secures agents at runtime. [s27, s4, s20] |
Tenet Security addresses a specific risk. An AI agent with legitimate access can follow instructions an attacker hid in data it reads. Tenet calls this attack Agentjacking. Its launch announcement describes malicious instructions embedded in emails, documents, logs or databases that steer agents within their authorized permissions.
Tenet sells to enterprise security teams. Its about page says the founders started the company to let security teams say yes to AI initiatives. SecurityWeek describes the aim as detecting and stopping dangerous agent behavior in real time.
Sources independent of Tenet describe the same gap. VentureBeat cites five surveys from January to June 2026 on how far enterprises trust their agents. In one, from Okta and Apprize360, 34% of organizations apply the same controls to agents as to humans. A Cloud Security Alliance research note on Agentjacking adds a policy gap. Cloud Security Alliance researchers write that most organizations lack explicit policies governing which data sources agents may query. [s26, s3, s12, s20, s24]
Tenet Security Platform covers discovery, detection and investigation, protection, management, and testing before production. The product page says Tenet maps every AI agent, MCP server and tool integration from traffic metadata. It also reconstructs each agent session, including identity, tool calls and intent.
Protection centers on what Tenet calls Agent-Side Simulation, which the company describes as patent pending. SecurityWeek reports that when Tenet’s sensor flags a suspicious action, the platform predicts the agent’s next move before it executes. The product page says on-device models run that simulation and that Tenet’s runtime layer stops harmful actions at the source.
Coverage depends on where the agents run. The FAQ says the Tenet Sensor covers endpoint agents through existing EDR and MDM infrastructure. For cloud agents, the FAQ names read-only trace logs or a 10-line gateway plugin for active defense. Tenet’s comparison page instead describes a sensor inside the agent runtime across endpoint and cloud, with no gateways. The company offers the platform self-hosted or as SaaS.
The product page says Tenet red-teams each agent in CI/CD against the OWASP Top 10 for agents and MITRE ATLAS. Tenet supplies the performance figures itself, including sub-30ms for active defense. [s1, s2, s12, s6, s4, s5]
Tenet positions its sensor against controls deployed outside the agent. Its comparison page argues that AI gateways cannot see MCP calls, local file reads and tool calls. The same page says EDR watches processes and files without seeing what agents do inside them.
Techspective writes that a growing list of funded startups is chasing the same problem. In a VentureBeat article on Agentjacking, CrowdStrike’s chief technology officer, Elia Zaitsev, says CrowdStrike is now securing agents at runtime.
Tenet also reports outside placements in its own announcements. Its blog says the Cloud Security Alliance’s May 2026 market map lists Tenet in three categories. Another post says Tenet joined the Claude Partner Network in August 2026. [s4, s25, s20, s28, s27]
Tenet Security emerged from stealth in June 2026 with $6 million in seed funding. Ctech reports that The Westly Group and MizMaa Venture led the round. SiliconANGLE describes Westly as the lead with MizMaa participating, and SecurityWeek names Westly as the lead.
Tenet does not name the customers in its cited legal-sector and Fortune 1000 deployment examples. It says a legal-sector enterprise with $1 billion in annual recurring revenue grew from two agent deployments to more than 20. Tenet says that took six months and that its platform blocked more than 10 attempted attacks there. In a Fortune 1000 proof of value, Tenet says it caught a runaway agent burning tens of thousands of dollars in tokens.
Outlets that relayed those figures attribute them to Tenet. Techspective writes that the numbers come from Tenet, and Ynetnews says the company did not disclose customer names. The about page quotes named security leaders under a “Trusted at the Frontier” heading, including Rick Scot of Elevate Textiles and Ravi Nori of GoPuff.
Tenet’s attack research has drawn wide press coverage. The Hacker News, Infosecurity Magazine, The Next Web and VentureBeat reported its June 2026 Agentjacking disclosure. VentureBeat also reported the GhostJacking attack Tenet demonstrated on the DEF CON 34 main stage on August 9. [s22, s13, s12, s25, s32, s3, s14, s16, s17, s20, s21]
Co-founders Barak Sternberg, the chief executive, and Nevo Poran, the chief technology officer, come from offensive security research. SecurityWeek reports that the two previously built Cisco’s AI Defense research team together. It adds that both are alumni of Unit 8200, the Israeli military intelligence unit.
The founders also built an earlier company. SiliconANGLE reports that they ran Wild Pointer, a cybersecurity company that reached seven-figure annual revenue with Fortune 500 customers. Techspective adds that they walked away from Wild Pointer to start Tenet.
Their research has a public record. SiliconANGLE reports that both have spoken at DEF CON and Black Hat. The Hacker News names Ron Bobrov, Barak Sternberg and Nevo Poran as the researchers behind Agentjacking. [s12, s13, s25, s14, s3]
Tenet Security Inc. owns and operates the company’s website, according to its terms of use. The FAQ claims full SOC 2 compliant status. The probed /trust path returned a 404 page, and the reviewed sources carry no audit report, so the claim is Tenet’s own statement.
The FAQ describes how the hosted service handles customer data. It says raw sensitive data exists only in memory and is redacted before storage in encrypted AWS-managed databases. It also describes encryption at rest and in transit, tenant isolation and rotating keys.
Tenet presents its testing as support for customers’ own compliance. Its how-it-works page lists the OWASP Top 10 for agents, NIST AI RMF and the EU AI Act as test frameworks. [s10, s6, s31, s5]
A closer look at the company's product strategy, measuring how defensible it is against market forces and examining the eight areas behind it.
press the advantage
Tenet Security sells software that customers run themselves, connected to their EDR tools, gateways and agent platforms. Tenet blocks agent actions where a customer installs its in-runtime sensor or gateway plugin. Discovery and trace reading run alongside the agents. The company says its detectors draw on millions of agent sessions and a honeypot network it runs. Its stated compliance credential is a SOC 2 claim. A customer leaving Tenet would remove its adapters, sensor and gateway plugin and rebuild its guardrail policies in another product.
| 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 | Tenet sells a software platform that the customer’s security team runs itself. The team manages adapters, members and security policies from a single sign-on dashboard. |
| Switching Cost How expensive leaving is for a customer: data portability, integrations, learned workflows, network effects, regulatory data residency. | 2/3 | Customers connect Tenet through adapters, a sensor that uses their EDR and MDM tools, and a gateway plugin. They tune its guardrails to their business logic, so leaving means removing those connections and rebuilding policies elsewhere. The cited record does not size the migration or show non-portable state. |
| Compliance Moat Whether certifications, liability acceptance, or audit trails block an easy replacement. | 1/3 | Tenet’s FAQ claims SOC 2 compliant status, an attestation a funded competitor can obtain through ordinary preparation. The probed /trust path returned a 404 page. No cited source names a regulation that requires this product or an authorization only Tenet holds. |
| Problem Complexity Whether the product requires ML, optimization, real-time systems, or years of specialized expertise. | 3/3 | Tenet simulates an agent’s next action with on-device models and blocks harmful actions in real time. It reconstructs sessions across operating-system, network, API and LLM activity. That is real-time ML and systems work, and the company claims sub-30ms for active defense. |
| Buyer Profile Whether buyers are SMB operators, mid-market IT teams, or regulated enterprises and governments with procurement gates. | 3/3 | Buyer profile scores the buyer class the product is built and sold for. Tenet sells to enterprise security leaders. Its published customer quotes come from a CISO at a Fortune 500 financial services company and a U.S. legal institution, both unnamed and on Tenet’s own pages. |
| Layer Whether the product is an end-user application, a platform with application features, or infrastructure other applications depend on. | 3/3 | Under scoring policy 1.7.0, a component in the execution path counts as infrastructure. Tenet describes a sensor that sits inside the agent runtime and, before an action executes, simulates where it leads and blocks the path if it is dangerous. Tenet reports one enterprise that scaled from two to more than twenty agent deployments while the sensor blocked real attacks. Where a customer deploys it, agent actions depend on that component. Discovery and trace reading also run out of band, and the deployment account is Tenet's own. That meets rung 3. |
| Proprietary Data, Content, or IP Whether the product accumulates datasets, content licenses, or IP that a rival cannot recreate from scratch. | 2/3 | Tenet says its detectors are trained on millions of live agentic sessions and updated from its Threat Lab and a honeypot network it runs. That is a vendor-held asset the cited record does not show as public. Tenet provides the training figures without identifying whose sessions it used. It describes Agent-Side Simulation as patent pending. |
Tenet Security builds for enterprise security teams whose organizations run AI agents in production. Its homepage frames the offer around coding agents, cloud agents, support agents and customer-facing agents. Its Claude Partner Network post lists coding agents such as Cursor, Codex, Windsurf and GitHub Copilot. It also lists cloud agents on Azure Foundry and Bedrock.
The customer quotes Tenet publishes point at large organizations, including financial services. One comes from a chief information security officer at a Fortune 500 financial services company. Another comes from a security head at a U.S. legal institution the page describes as $1Bn+. Neither quote names the organization.
SecurityWeek reports that the seed round funds expansion of Tenet’s North American go-to-market operations.
Tenet’s central technique predicts where an agent action leads before it runs. The product page says on-device models simulate the path. The how-it-works page says the simulation runs in a sandboxed environment. SecurityWeek reports that the simulation starts when the sensor flags a suspicious action.
Tenet says its detection draws on data it gathers itself. Its comparison page claims training on millions of live agentic sessions across more than 1,000 agent profiles. It also claims more than 100 detectors updated from Tenet’s Threat Lab and a global honeypot network. Those figures come from the company.
Tenet’s research shows the attack class its product targets. In Agentjacking, attackers hide instructions in data an AI agent reads. The Hacker News reports that Tenet tested Agentjacking against more than 100 organizations with an 85% exploitation success rate. VentureBeat and the Cloud Security Alliance call that work proof-of-concept research in controlled testing. Neither reports confirmed exploitation at scale.
Tenet sells through a sales-led motion that opens with an assessment. Its homepage offers a full agent risk assessment in under three hours, and its site offers a demo request. SecurityWeek reports that the seed money funds North American go-to-market expansion.
Tenet’s research disclosures have drawn coverage from several security and technology publications. The Hacker News, Infosecurity Magazine, The Next Web and VentureBeat covered Agentjacking in June 2026. VentureBeat covered the GhostJacking demonstration Tenet gave on the DEF CON 34 main stage on August 9.
Tenet describes early deployments without naming customers, and outlets relaying them attribute the figures to Tenet. The company says it joined the Claude Partner Network in August 2026.
Tenet publishes no price list. The probed /pricing path returned a 404 page. The homepage routes buyers to a demo request or an agent risk assessment. The reviewed sources state no pricing unit or tier.
Customers run Tenet as SaaS or self-hosted, according to the how-it-works page. Setup uses lightweight connectors with no heavy SDKs, the company says. After integration, teams use a dashboard behind single sign-on to manage adapters, invite members and set security policies.
Deployment differs between endpoint and cloud agents. The FAQ says endpoint coverage runs through the Tenet Sensor using existing EDR and MDM infrastructure. For cloud agents, the FAQ names read-only trace logs or a 10-line gateway plugin for active defense. Tenet’s comparison page describes a sensor inside the agent runtime across endpoint and cloud, and the reviewed sources do not say which route each cloud deployment uses. The company describes its adapters as out-of-band connectors.
Tenet Security Inc. operates the company’s website, according to its terms of use. Tenet states in its FAQ that it complies with SOC 2. The reviewed sources contain no audit report.
The FAQ describes data handling for the hosted service. It says raw sensitive data exists only in memory and is redacted before it reaches encrypted AWS-managed databases. It also describes encryption at rest and in transit, tenant isolation and rotating keys.
Tenet connects to the security tools and agent platforms a customer already runs. The how-it-works page lists EDR, SSE network visibility and GitHub repository activity among its connections. The Claude Partner Network post lists coding agents such as Cursor, Codex, Windsurf and GitHub Copilot. It adds cloud agents on Azure Foundry and Bedrock.
Tenet added Claude-specific coverage in August 2026. The announcement describes discovery across Claude agents, users and configurations, and inference hooks for Claude in beta. Tenet’s testing maps to the OWASP Top 10 for agents and MITRE ATLAS.
Tenet publishes tools in public repositories on GitHub. One, agent-jackstop, hardens Cursor and Claude Code against prompt injection through untrusted tool output. Another scans for artifacts of the Cline agent supply-chain compromise.
Barak Sternberg, the chief executive, and Nevo Poran, the chief technology officer, co-founded Tenet after offensive security research work. SecurityWeek reports that they previously built Cisco’s AI Defense research team together and that both are Unit 8200 alumni. SiliconANGLE reports that they earlier ran Wild Pointer, a cybersecurity company with seven-figure annual revenue and Fortune 500 customers.
Their research carries public standing. SiliconANGLE reports that both have spoken at DEF CON and Black Hat. VentureBeat reports Tenet’s GhostJacking demonstration on the DEF CON 34 main stage.
Tenet’s advisors and angel investors include security executives. SiliconANGLE names two advisors, David Schwed, formerly CISO at Robinhood, and Rick Scott, formerly CISO at BNY Mellon. Ctech and Ynetnews name Dazz founder Tomer Schwartz and former Coralogix chief executive Lior Tal among the angel investors.
| Id | Source | Tier | Accessed |
|---|---|---|---|
| f1 | Tenet Security: product | official | 2026-09-23 |
| f2 | Fortune: Term Sheet deals, June 17, 2026 | press | 2026-09-23 |
| f3 | Ctech: Tenet raises $6 million in Seed funding to address risks of autonomous AI agents | press | 2026-09-23 |
| f4 | AI Defense Matrix Catalog mapping | other | 2026-09-23 |
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.