Radiant Security

• Well-funded AI SOC rivals such as Dropzone AI, 7AI, Prophet Security, and Torq sell near-identical triage and response outcomes. If buyers cannot tell them apart in evaluation, price competition compresses Radiant's position.
• Investigation quality depends on outside-lab generative models, with no proprietary model evidenced in public sources, that every competitor can license. If model capability converges, differentiation narrows to integrations and embedded logic that incumbents can replicate.
• The newest disclosed raise is the November 2023 Series A. If no later round or independently reported revenue milestone appears, the traction story depends on vendor-published customer references and claims.
• SIEM, SOAR, and MSSP platforms, plus Microsoft and Google, can bundle AI triage inside ecosystems buyers already license, turning Radiant's core capability into a platform feature.

Radiant sells into the security operations center, the team responsible for triaging and investigating the alerts a customer's detection stack raises. The company grounds the pain through a published MSSP customer, Single Point of Contact, whose team reached 100% alert coverage and multimillion-dollar savings after replacing manual triage with automation.

Demand for this outcome is broad and growing. An independent 2025 AI SOC market landscape report places Radiant among named vendors, and MSSP adoption shows buyers are actively shopping the category rather than being argued into it. The buyer can locate the budget line as SOC automation against existing SIEM, SOAR, and managed-detection spend.

The timing rests on a concrete enabler. Capable large language models arrived in 2022 and 2023, which made per-alert reasoning practical, and Radiant came out of stealth at Black Hat in 2023 to build on that shift. The window risk is platform absorption rather than absent demand. [s1, s5, s7]

The Radiant AI SOC Platform covers four jobs in one place, alert triage, investigation, response, and log management. Its stated differentiator is generative triage logic that builds a custom investigation plan for each alert rather than executing pre-built playbooks, which the company positions against rivals that rely on fixed playbooks.

Coverage breadth is the second claim. Radiant markets a large connector set, counted at 100-plus on its homepage and 120-plus on its about page, assembled as modular building blocks per investigation, so the platform can pull context from any tool in the stack rather than a fixed set per alert type. Built-in log management lets it ingest and analyze events without a separate SIEM, which the company frames as avoiding the SIEM cost.

These are real capabilities with external validation through a customer case study and a vendor award. They are also feature-level claims that a well-funded competitor can pursue, which is why the durable question is distribution and proprietary context rather than the feature list itself. [s2, s5, s7]

Radiant competes on two fronts, specialist AI SOC rivals and platform incumbents. Dropzone AI, 7AI, Prophet Security, and Torq all market AI SOC agents or platforms in near-identical language, and SIEM, SOAR, and MSSP platforms can ship comparable AI triage inside ecosystems buyers already license.

The company's stated edges are coverage breadth and being early. Radiant argues that starting before competitors gave it one of the larger integration libraries in the market, and that generative triage logic removes the alert-type ceiling that fixed-playbook tools hit. An MSSP audience values breadth because it manages many heterogeneous client environments.

No proprietary data asset appears in the public record to anchor a durable moat. Public sources evidence no proprietary model, so the investigation engine presumably runs on models any rival can license, and the defensible layer, if one forms, would be the integration depth and embedded SOC logic that a buyer comes to rely on, not the underlying model. [s2, s5, s7]

Radiant's public traction now reads as a roster rather than a single reference. The MSSP Single Point of Contact remains the most detailed case study, and the customers page adds vendor-published case studies for Rehrig Pacific, a $500M manufacturer that moved its SOC in-house from an MSSP, for Spellman High Voltage, and for the MSSP RFA. A Kyowa Kirin teaser, testimonials from Brian Rowe and Rodney Stewart, the 2026 vendor award, and the AI SOC landscape mention round out the record.

The funding timeline sharpens the question. The $15 million Series A closed in November 2023, led by Next47 with Lightspeed, Acrew, and others, and no later round has been disclosed in the more than two years since. Reputable backing supports an inference of some undisclosed traction, yet the reference depth remains vendor published rather than independently reported.

The motion appears to run through direct sales and MSSP channels rather than a published self-serve path. For an MSSP buyer managing many client environments, the broad integration set is the practical hook, and the Rehrig Pacific and Spellman stories extend the published motion to enterprises that bring SOC work in-house. [s3, s4, s5, s7, s9, s10, s11, s12]

Radiant's leadership carries verifiable security pedigree. Press describes the leadership as a team of former Imperva and Exabeam executives, prior builds in the same security-operations domain rather than adjacent experience, with founders Shahar Ben-Hador as CEO and Barry Shteiman as CTO named as security veterans.

The leadership bench is identifiable but thin in public detail. The about page lists a VP of Product and a VP of Engineering alongside the founders, and the company reports roughly 60 employees, a size consistent with a Series A team building rather than scaling a large go-to-market motion.

The team signal is strong on domain fit and weaker on independent recognition. The leadership team's former Imperva and Exabeam standing is the load-bearing credential, corroborated by press, while public publications, patents, or sustained community standing do not appear in the record reviewed. [s3, s6]

Radiant publishes an inspectable security attestation. Its trust center at trust.radiantsecurity.ai hosts a SOC 2 Type II report available for download, alongside documented infrastructure, organizational, and product security controls, which is table-stakes evidence for a platform that reads across a customer's security tools.

The platform's access posture raises the procurement bar. Because the AI ingests logs and reaches into a customer's SIEM, EDR, identity, and cloud tools to investigate, retention and model-usage terms still come through contract review even with the SOC 2 in hand.

No ISO 27001 or FedRAMP authorization appears in the public record reviewed. For an MSSP or enterprise SOC buyer, the published SOC 2 Type II and control documentation meet the baseline, while higher assurance for regulated buyers would require collateral not yet evident. [s8, s2]

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

Radiant targets the security operations center buyer, with MSSPs as a visible segment. A detailed published case study features an MSSP, Single Point of Contact, that manages many client environments and values coverage across heterogeneous stacks.

The segmentation logic favors breadth-sensitive buyers. An MSSP triaging alerts for many tenants cares about the integration count and the ability to handle any alert type, which is the value Radiant leads with, so the published reference and the positioning align.

Radiant's AI advantage claim is generative triage logic. Rather than executing pre-built playbooks, the platform builds a custom investigation plan per alert, reusing a plan when the alert is known and constructing one when it is new, which the company frames as removing the alert-type ceiling fixed-playbook tools hit.

The integration layer is the second capability. Radiant assembles a large connector set, counted at 100-plus on its homepage and 120-plus on its about page and a vendor award writeup, as modular building blocks per investigation, so it pulls context from any tool in the stack. Built-in log management ingests events without a separate SIEM.

These capabilities are real and externally described, but they are reproducible engineering rather than a proprietary asset. Public sources describe generative investigation logic without evidencing a proprietary model or exclusive dataset, so the reasoning layer is presumed replicable, which is the structural ceiling on the AI advantage.

Radiant's go-to-market runs through direct, demo-led sales rather than a published self-serve path, with MSSPs appearing as customers and deployment operators rather than a documented reseller channel. The site routes buyers to a booked demo, and a flagship reference is an MSSP, which fits a high-touch motion for a product that reads across a customer's security tools.

The published proof has widened from one case study to a roster. The SPoC MSSP remains a detailed reference, and the customers page now serves vendor-published case studies for Rehrig Pacific, a $500M manufacturer that moved its SOC in-house from an MSSP, for Spellman High Voltage, and for the MSSP RFA, with a Kyowa Kirin teaser alongside the 2026 vendor award and the AI SOC landscape mention. The outcome figures in these writeups are customer and vendor supplied rather than independently reported.

The one published price signal is for logging. A 2026 award writeup lists $23 per terabyte, while Radiant's own platform page markets unlimited log ingestion for a fraction of SIEM cost without publishing a rate, published signals that do not fully reconcile. The platform page contrasts its approach with usage-based competitors that it says lead to unexpected bills, and describes its own model as predictable and transparent by security use case, which signals negotiated deals over a self-serve list price.

The stated pricing philosophy targets the MSSP and enterprise buyer that wants budget predictability across many alert types. Beyond the logging signals, the unit Radiant charges by for the core platform cannot be confirmed from public sources.

Radiant delivers software the customer operates rather than a managed service. The case study describes the customer's own SOC team gaining productivity, with Radiant automating triage and ticket creation, so the analysts retain accountability for verdicts.

The deployment is fast by the customer's account. The case study describes onboarding across tenants as quick and easy, with Radiant plugging into the MSSP's ticketing system and becoming its go-to tool for incident handling, which points to a light onboarding for an out-of-path overlay.

Radiant publishes an inspectable attestation. Its trust center hosts a SOC 2 Type II report for download, with documented infrastructure, organizational, and product security controls, which is the baseline assurance a platform reading across a customer's tools needs.

The access posture keeps procurement involved. Because the AI ingests logs and reaches into a customer's SIEM, EDR, identity, and cloud tools, retention and model-usage terms remain for the buyer to settle in contracting even with the SOC 2 in hand. No ISO 27001 or FedRAMP authorization appears in the record reviewed.

Radiant positions as a layer over a customer's existing stack rather than a replacement. The 100-plus integrations into SIEM, EDR, cloud, identity, and other tools are the ecosystem play, and built-in log management lets it optionally displace some SIEM ingestion.

The ecosystem position cuts both ways. Reaching into the same consoles a buyer already owns means the console owners, SIEM, SOAR, EDR, and MSSP platforms, sit closest to bundling a comparable AI analyst, which is the structural pressure on Radiant's independence.

Radiant's leadership carries domain-matched pedigree. Press describes the leadership as a team of former Imperva and Exabeam executives, and the named founders, Shahar Ben-Hador as CEO and Barry Shteiman as CTO, are described as cybersecurity veterans, executive experience inside the security industry rather than adjacent fields.

The bench is identifiable but lightly documented in public. The about page lists a VP of Product alongside the founders, and the company reports roughly 60 employees, consistent with a Series A team still building toward a larger go-to-market motion.