Radiant Security
AI SOC platform triaging every alert type that reaches the SOC, with analyst confirmed response actions and integrated log management. Cribl acquired its technology in August 2026.
Radiant Security builds an AI SOC platform that triages and investigates the alerts a security team's tools raise, across endpoint, identity, email, network and cloud, and pairs it with integrated log management. In August 2026 Cribl acquired the technology assets and intellectual property of Radiant's AI SOC product, which it plans to adapt as an application on its telemetry data platform. Radiant still presents and sells the product on its own site and help center.
Alerts arrive through data connectors or a custom alerts webhook and are triaged and investigated automatically, and each case carries an incident specific response plan. Response actions run from the case view across CrowdStrike, Microsoft 365 Defender, SentinelOne, Microsoft 365, Okta, Google Workspace, Mimecast, Proofpoint, Netskope and KnowBe4.
They isolate devices, disable users, reset passwords, end sessions, delete messages and block IPs, URLs and domains. An analyst selects the targets and confirms each action, and many can be reversed with one click. Audit logs record response actions, verdicts and status changes. Log management stores the customer's logs with search and retention settings, and SSO runs through Okta, Google or Microsoft Entra ID.
Radiant publishes no pricing, model providers or hosting regions.
Vendor details
Canonical URL
https://radiantsecurity.ai
Category
Security / SOC agent
Subcategory
AI SOC analyst
Funding status
Venture backed, raised a $15 million Series A. On 19 August 2026 Cribl announced it had acquired the technology assets and intellectual property of Radiant's AI SOC product.
Company status
acquired
Use cases & customers
Primary use cases
Target customers
Deployment options
Integrations
Radiant ingests alerts from existing detection tools through data connectors or a custom alerts webhook and triages every alert type that reaches the SOC, with integrated log management alongside. Response actions, each confirmed by an analyst, run in CrowdStrike, Microsoft 365 Defender, SentinelOne, Microsoft 365, Okta, Google Workspace, Mimecast, Proofpoint, Netskope and KnowBe4.
In practice
Your alert stream spans endpoint, identity, and cloud and you want one triage layer for all of it. Radiant triages every alert type that reaches the SOC, across endpoint, identity, email, network and cloud.
Your team wants to search its logs and alerts in the same place it triages them. Radiant's integrated log management stores the logs with search, query and retention settings, and alerts sent through the custom webhook land in indexes the team can search and investigate.
You want response, not just prioritized alerts. Radiant investigates each alert automatically and builds an incident specific response plan, and an analyst confirms each response action, many of which can be reversed with one click.
Sources & related URLs
Research sources
Agentic Index coverage score
6.5 / 14 capabilities · 46%
| Integrations & Tool Calling | Full |
|---|---|
|
Data connectors ingest from the customer's tools in categories that include authentication, cloud access security brokers, cloud, endpoint and network logs, email infrastructure and IAM, ICS and OT, messaging apps, password managers, SIEM tools, ticketing systems, security service edge, API protection and an S3 connector, and outgoing webhooks send data on. Response actions run from the case view in CrowdStrike, Microsoft 365 Defender, SentinelOne, Microsoft 365, Okta, Google Workspace, Mimecast, Proofpoint, Netskope and KnowBe4, isolating devices, disabling users, resetting passwords, ending sessions, deleting messages, removing external forwarding rules and blocking IPs, URLs, domains and files. They also start full disk scans in SentinelOne and enroll users in KnowBe4 phishing training. SourceRadiant Security, help.radiantsecurity.ai and /radiant-cases/radiant-cases/response-actionsread 2026-10-06 |
|
| Workflow Orchestration | Partial |
|
Each alert runs through the product's own triage, investigation and case sequence, and response actions launch from the case, on one artifact or in bulk across selected users, IPs and URLs. The audit log mentions automated playbooks, but Radiant describes no branching, multi agent handoff or automation the customer configures. SourceRadiant Security, help.radiantsecurity.ai/radiant-cases/radiant-cases/response-actions-in-cases and /log-management/audit-logs/introduction-to-audit-logsread 2026-10-06 |
|
| Knowledge Grounding & RAG | Partial |
|
Integrated log management stores the customer's logs with search, query and retention settings, and alerts sent through the custom webhook land in its parsed and alert indexes, where teams can search and investigate them. Radiant says the platform continuously learns the environment, its tooling, behavior and activity to boost accuracy, but it does not say how investigations draw on what it has learned or on the customer's own documentation. SourceRadiant Security, radiantsecurity.ai, help.radiantsecurity.ai/log-management/log-search-and-query and /radiant-connectors/ingestion-methods/custom-alerts-webhookread 2026-10-06 |
|
| Human Oversight & Guardrails | Full |
|
Response actions launched from a case run when an analyst selects the targets, clicks the action and confirms the list of affected artifacts in a confirmation step, and many can be reversed with one click from the action history through the inverse action, such as enabling a user again. Hard deletes and password resets cannot be undone. The platform produces incident specific response plans for that analyst to execute. The audit log separates analyst actions from automated platform actions and playbooks, and Radiant does not say which actions, if any, run without that confirmation. SourceRadiant Security, help.radiantsecurity.ai/radiant-cases/radiant-cases/response-actions-in-cases, /response-actions and /log-management/audit-logs/introduction-to-audit-logsread 2026-10-06 |
|
| Security, Identity & Governance | Partial |
|
SAML 2.0 SSO works with Okta, Google and Microsoft Entra ID and is enforced for all users once set, with an emergency bypass through support. Every user is assigned the Admin role, and Radiant names no SCIM provisioning. No compliance attestation is published, and a trust center sits at trust.radiantsecurity.ai. SourceRadiant Security, help.radiantsecurity.ai/manage-radiant/organization-settings/security/set-up-single-sign-on-ssoread 2026-10-06 |
|
| Observability & Auditability | Partial |
|
Audit logs record who took action, what changed and when, with before and after values, across response actions, alert verdict changes, allow and deny list changes, deleted notes and data connector changes, and they tell analyst actions apart from automated platform actions. Configuration and administrative changes are not yet logged, and Radiant describes no record of the AI's investigation steps and reasoning. SourceRadiant Security, help.radiantsecurity.ai/log-management/audit-logs/introduction-to-audit-logsread 2026-10-06 |
|
| Memory & State Persistence | Not documented |
|
The platform continuously learns about the environment, its tooling, behavior and activity to boost accuracy, by Radiant's account, but Radiant describes no memory store, what it would keep, its scope or its lifetime. SourceRadiant Security, radiantsecurity.ai and help.radiantsecurity.airead 2026-10-06 |
|
| Deployment & Data Residency | Not documented |
|
How the product is hosted is not described, and no hosting region, choice of region or customer environment option is published. SourceRadiant Security, radiantsecurity.ai and help.radiantsecurity.ai; trust.radiantsecurity.ai/resourcesread 2026-10-06 |
|
| Prebuilt Agents, Templates & Packs | Partial |
|
One AI SOC analyst with incident response and log management around it covers many alert types out of the box, triaging every security alert across endpoint, identity, email, network and cloud. These are parts of one product, with no separate prebuilt agents or templates to adopt. Named customers include Nutcracker Therapeutics, Spellman High Voltage Electronics, Kyowa Kirin and Second Wave Delivery Systems. SourceRadiant Security, radiantsecurity.ai and help.radiantsecurity.airead 2026-10-06 |
|
| Triggers & Channel Coverage | Full |
|
Alerts arriving through data connectors or the custom alerts webhook are triaged and investigated automatically, with no analyst starting the work. Webhook alerts are posted as JSON with a token, enter the triage pipeline and become searchable within several minutes. SourceRadiant Security, help.radiantsecurity.ai/radiant-connectors/ingestion-methods/custom-alerts-webhookread 2026-10-06 |
|
| Model Flexibility & Routing | Not documented |
|
The models and providers behind Radiant's AI are not disclosed, and no model choice is offered. SourceRadiant Security, radiantsecurity.ai and help.radiantsecurity.airead 2026-10-06 |
|
| APIs, SDKs & MCP Extensibility | Full |
|
The custom alerts webhook has a published endpoint, a token sent in the authorization header and a JSON body carrying timestamp, alert ID and raw alert fields, for sending alerts from sources without a dedicated connector. Radiant has announced an API for automating operational work, but no API reference beyond the webhook is published. SourceRadiant Security, help.radiantsecurity.ai/radiant-connectors/ingestion-methods/custom-alerts-webhookread 2026-10-06 |
|
| Testing, Debugging & Optimization | Not documented |
|
Claimed results are triage with the quality of a team's best analyst at any scale and response times cut from days to minutes, both Radiant's own figures. Radiant describes no evaluation harness, scored test cases, quality gate or optimization loop the customer runs. SourceRadiant Security, radiantsecurity.ai and help.radiantsecurity.airead 2026-10-06 |
|
| Browser & Computer Use | Not documented |
|
The AI works through data connectors and response actions inside the customer's security and identity tools. Radiant describes no use of a browser, desktop or remote computer. SourceRadiant Security, radiantsecurity.ai and help.radiantsecurity.ai/radiant-cases/radiant-cases/response-actionsread 2026-10-06 |
|
The Agentic Index coverage score grades every vendor Full, Partial or Not documented against the same 14 buyer facing capabilities, from public evidence only. Each capability links to how all vendors in the index score on it. How this evidence is graded
Product updates
Cribl has acquired the technology assets and intellectual property behind Radiant Security's AI-native SOC product. The acquisition covers Radiant's software for autonomous alert triage, investigation, and resolution, which Cribl plans to adapt as an application on its telemetry data platform.
Bears on: Funding / partnership
View sourcePricing
Contact sales; enterprise contracts with optional integrated log management
enterprise contract
Included quota
Platform contract covering adaptive triage across all alert types, with integrated log management as an option; no public tiers.
What is public
Nothing numeric; the adaptive triage and integrated log management positioning are public.
Billing mechanics
Enterprise platform contracts through sales for adaptive triage, with integrated log management offered as a SIEM cost alternative.
Cost watchouts
Integrated log management changes the cost equation: it can offset a separate SIEM but adds storage scope to the Radiant contract. Compare against keeping the existing SIEM plus a pure overlay analyst.
Variable cost rationale
Enterprise platform licensing; integrated log management may scale with data retained, but no usage metering is documented, so exposure sits at low.
Additional watchouts
Cribl announced in August 2026 that it acquired the technology and IP of Radiant's AI SOC product; confirm contract continuity, support and the product roadmap before committing.
Overage / add-ons
No public metering documented; log management scope would likely affect cost.
Sales call required
Yes, required for paid access
Free / trial
Enterprise evaluations through sales; no self serve trial
Lowest paid plan
None public; enterprise contract only
Commercial notes
Venture backed on a $15 million Series A. Cribl announced the acquisition of the AI SOC product's technology assets and IP on 19 August 2026.
Key ambiguities
Nothing numeric is public, and whether log management is bundled or priced separately is not documented.
Related vendors
- 7AI — Agentic SOC from the Cybereason founders: domain specialized agents…
- Abnormal AI — Behavioral AI email security with three named autonomous agents: a…
- AirMDR — AI native MDR whose AI analyst investigates every alert and writes…
- Airrived — Agentic OS that unifies SOC, GRC, IAM, vulnerability management, IT,…
- Andesite — Human-AI SOC platform where customers build their own agents and…
- Anvilogic — Agentic security operations platform that works on top of existing…
Alternatives to Radiant Security
The closest documented capability profiles to Radiant Security among security and SOC agents tracked by Agentic Index, ordered by similarity on the same 14 point evidence the rankings use. No vendor pays for placement.
- AirMDR6.0 / 14Fuller documented coverage on Security, Identity & Governance
- Intezer7.0 / 14Adds documented Memory & State Persistence
- Darktrace7.5 / 14Adds documented Deployment & Data Residency
- Linx Security6.5 / 14Fuller documented coverage on Knowledge Grounding & RAG and Observability & Auditability
- Okta8.5 / 14Adds documented Deployment & Data Residency
- Token Security6.5 / 14Fuller documented coverage on Workflow Orchestration and Knowledge Grounding & RAG
Similarity is computed from each vendor's Agentic Index coverage score evidence, axis by axis, not from the totals. How this evidence is graded