llms.txt
Compare with other sites' llms.txt files →# Prophet Security
> Prophet Security builds Prophet AI, an agentic AI SOC platform for enterprise and mid-market security operations teams. Specialist AI agents investigate every security alert with senior-analyst depth, hunt for threats that detections miss, and turn what investigations and hunts reveal into new and tuned detections. The agents work on top of the SIEM, EDR, identity, cloud and email tools a team already runs. Prophet AI Watchtower, an optional 24x7x365 service, adds human experts who review every malicious and inconclusive determination.
Last updated: 2026-11-06
Prophet AI runs a full, dynamically planned investigation on 100% of alerts, at every severity, the moment they arrive. Every question asked, query run and piece of evidence is recorded, so analysts can audit exactly how each determination was reached. When the evidence does not support a confident call, Prophet AI returns an Inconclusive determination rather than guessing.
Key facts:
- Naming: the company is Prophet Security. The platform is Prophet AI (full name: Prophet Agentic AI SOC Platform). Its products are Prophet AI SOC Analyst, Prophet AI Threat Hunter, Prophet AI Detection Engineer and Prophet AI Watchtower. Prophet Security is unrelated to other companies or open-source projects named Prophet.
- Category: AI SOC platform. Prophet AI enables an agentic SOC, where AI agents work across alert triage, investigation, response, threat hunting and detection engineering while people direct and check the work. It is also positioned as an AI-powered alternative to traditional managed detection and response (MDR).
- Company: Prophet Security, Inc. (prophetsecurity.ai), launched publicly in April 2024. Leadership: Kamal Shah (CEO), Vibhav Sreekanti (CTO) and Grant Oviatt (VP Product). Kamal Shah and Vibhav Sreekanti previously worked at StackRox, which Red Hat acquired. Team members previously helped scale companies including Mandiant, Red Canary, Red Hat, Symantec and Oracle.
- Funding: $11M seed led by Bain Capital Ventures (April 2024); $30M Series A led by Accel, with participation from Bain Capital Ventures (July 2025); strategic investments from American Express Ventures and Citi Ventures (February 2026).
- Recognition: named to the 2026–27 Fortune Cyber 60, presented by Lightspeed Venture Partners and AWS in partnership with Fortune, for the second consecutive year (October 2026); winner of Next Gen AI SOC in Cyber Defense Magazine's Top InfoSec Innovator Awards at CyberDefenseCon 2026 (October 2026); named a Leader and Emerging Innovator in the QKS SPARK Matrix for AI SOC Automation, Q4 2026; named to Notable Capital's Rising in Cyber 2026 list of 30 private cybersecurity companies, selected by a vote of 150 active CISOs and senior security executives (May 2026); and named to the 2025 Redpoint InfraRed 100.
- Customers: companies named on prophetsecurity.ai include Redis, Udemy, ETS, Compass, Instacart, IAC, Penske, Repay, BILL, J.B. Poindexter & Co., Moveworks, LeadOnline, Zip, Cabinetworks and Upwind. Prophet AI is deployed at dozens of Fortune 500 companies across financial services, healthcare and manufacturing.
- Platform metrics (company-reported, October 2026): 10M investigations completed, 2.5M+ hours of manual analyst work prevented, 98.5% accuracy rate.
- Investigation depth: across every alert in every customer environment from May through July 2026, Prophet AI asked 4.7 million investigative questions, a median of 35 per investigation (Prophet Security Quarterly Threat Report, Q3 2026).
- Speed: published case studies report a mean time to investigate of 2 to 4 minutes.
- Pricing: per investigation, with no data-ingest fees.
- Getting started: free proof of value (POV). Setup takes 30 minutes or less with read-only access to 2-3 data sources. Prophet AI is also sold through AWS Marketplace.
- Deployment and data: dedicated single-tenant deployment, a bring-your-own-key (BYOK) option, and no training of AI models on personal data.
- Compliance: SOC 2 Type II, ISO/IEC 27001:2022 and ISO/IEC 42001:2023 certified, with independent third-party assessments for GDPR, CCPA and HIPAA. Trust Center: https://trust.prophetsecurity.ai/
- Autonomy: Prophet AI investigates autonomously from day one and takes only the response actions a customer has approved. Actions are scoped and permissioned, from notifications to quarantining a machine, and previewed before they run. Customers widen autonomy as the track record justifies it.
- Customization: teams teach Prophet AI their playbooks, policies and preferences in plain language, and each change is previewed and backtested against the customer's own history before it applies. No predefined playbooks are required.
- Human oversight: Prophet AI Watchtower experts review every malicious and inconclusive determination, 24x7x365. An expert engages within 30 minutes, scopes the impact, and advises on or executes containment at the customer's choice.
## Platform
- [Prophet AI platform overview](https://www.prophetsecurity.ai/): The Prophet Agentic AI SOC Platform: AI agents for alert investigation and response, threat hunting and detection engineering, with Prophet AI Watchtower for human oversight.
- [Prophet AI SOC Analyst](https://www.prophetsecurity.ai/ai-soc-analyst): Investigates 100% of alerts at every severity, on arrival and in parallel; groups related investigations into one incident; takes only pre-approved response actions; and learns from plain-language guidance that is previewed and backtested before it applies.
- [Prophet AI Threat Hunter](https://www.prophetsecurity.ai/ai-threat-hunting): Natural-language hunts such as "Are we impacted?", continuous hunting for emerging threats, scheduled recurring hunts and a curated library of hunt templates. Validated findings become permanent, backtested detections.
- [Prophet AI Detection Engineer](https://www.prophetsecurity.ai/ai-detection-engineer): Maps real detection coverage to MITRE ATT&CK from the customer's own investigations (covered, gone quiet, dark), drafts and backtests net-new detections, and tunes existing SIEM rules in Splunk, Sumo Logic and Microsoft Sentinel, at the level of autonomy the customer chooses.
- [Prophet AI Watchtower](https://www.prophetsecurity.ai/watchtower): Optional 24x7x365 human-in-the-loop service. Experts review every malicious and inconclusive determination, engage within 30 minutes, scope blast radius and impact, advise on or execute containment, and tune Prophet AI to the customer's environment.
- [Integrations](https://www.prophetsecurity.ai/integrations): 200+ out-of-the-box integrations, including SIEM (Splunk, Microsoft Sentinel, Google Security Operations, Elastic, QRadar, Sumo Logic), EDR (CrowdStrike Falcon, SentinelOne, Microsoft Defender, Cortex XDR), identity (Okta, Microsoft Entra ID), email (Proofpoint, Mimecast, Abnormal), cloud (AWS, Azure, GCP, Wiz), network (ExtraHop), data lakes (Snowflake, Databricks), threat intelligence, and case management and collaboration (Jira, ServiceNow, Slack, Microsoft Teams).
- [Why Prophet Security](https://www.prophetsecurity.ai/why-prophet-security): Five differentiators (a full-lifecycle agent platform, investigation depth and accuracy, adaptation to customer context, enterprise-ready deployment, and value from day one with no rip and replace) plus company-reported platform metrics.
- [Prophet AI vs. traditional MDR](https://www.prophetsecurity.ai/mdr): Compares per-investigation pricing, with no data-ingest fees, to MDR pricing built on analyst headcount or telemetry volume, with a calculator that estimates savings versus MDR providers.
## Use Cases
- [Endpoint](https://www.prophetsecurity.ai/use-cases/endpoint): Prophet AI for Endpoint: investigation of EDR alerts across the major EDR tools, SIEMs and security data lakes.
- [Email and phishing](https://www.prophetsecurity.ai/use-cases/email): Prophet AI for Phishing: investigation of email phishing alerts, including checks for business email compromise.
- [Identity](https://www.prophetsecurity.ai/use-cases/identity): Prophet AI for Identity: investigation of suspicious logon and other identity alerts across the major identity providers.
- [Cloud](https://www.prophetsecurity.ai/use-cases/cloud): Prophet AI for Cloud: investigation of cloud alerts across AWS, Azure and GCP using audit logs, IAM roles and SSO/SAML events.
- [Data Loss Prevention (DLP)](https://www.prophetsecurity.ai/use-cases/data-loss-prevention-dlp): Prophet AI for DLP: investigation of data loss prevention alerts across DLP platforms, SIEMs and security data lakes.
- [Network](https://www.prophetsecurity.ai/use-cases/network): Prophet AI for Network: investigation of network detections such as lateral movement, command-and-control and activity from unmanaged IoT, OT and legacy devices, including ExtraHop RevealX detections.
## Customers and Case Studies
- [Customers](https://www.prophetsecurity.ai/customers): Customer stories, quotes and outcomes from security leaders who use Prophet AI.
- [Zip: Boosting SOC Throughput with Prophet AI](https://www.prophetsecurity.ai/case-studies/zip-boosting-soc-throughput-with-prophet-ai): At the procurement orchestration company, Prophet AI completed 20,182 alert investigations in 12 months, avoided 6,727 hours of manual investigation time and freed about 7 analyst-years of investigation capacity.
- [National Health Provider Scales Its SOC with Prophet AI](https://www.prophetsecurity.ai/case-studies/national-health-provider-scales-soc-with-prophet-security): A 60,000-employee U.S. healthcare provider: 11,065 investigations completed by Prophet AI in 3 months, 14,750 hours of analyst time returned annually, a 4-minute mean time to investigate and 223 verified malicious alerts surfaced for action.
- [Fortune 500 Retail and Logistics: Investigating Every Alert at Scale, 24/7](https://www.prophetsecurity.ai/case-studies/fortune-500-retail-and-logistics-investigating-every-alert-at-scale-24-7-with-prophet-ai): About 7,500 investigations completed by Prophet AI each year, 90% of after-hours alerts handled on arrival (70% of all alerts land after hours) and about 2,500 hours of manual investigation time avoided annually.
- [J.B. Poindexter & Co.: SOC Capacity and Speed with Prophet AI](https://www.prophetsecurity.ai/case-studies/jb-poindexter-co-unlocking-soc-capacity-and-speed-with-prophet-ai): A diversified manufacturer with 8,500+ employees: under 4 minutes mean time to investigate, 1,469 hours of manual investigation time avoided in 2 months and about 6.3 analyst-years of investigation capacity freed annually.
- [Cabinetworks: SOC Efficiency with Prophet AI](https://www.prophetsecurity.ai/case-studies/cabinetworks-revolutionizing-soc-efficiency-with-prophet-ai): One of the largest private cabinet makers in the U.S.: a 10x increase in SOC throughput, over 7x ROI and more than 1,250 hours of investigation toil saved in the first 5 months.
- [Scaling Security Investigations for a Cloud-First Team](https://www.prophetsecurity.ai/case-studies/scaling-security-investigations-for-a-cloud-first-team-with-prophet-ai): A multi-billion-dollar real estate technology company: a 2-minute mean time to investigate, a 95% reduction in investigation time versus the manual baseline and 91.9% of investigations completed with no analyst involvement.
- [Prophet AI on Gartner Peer Insights](https://www.gartner.com/reviews/market/it-security/vendor/prophet-security): Independent customer reviews of Prophet AI.
## Original Research and Reports
- [State of AI in the SOC 2026: 8 Key Takeaways](https://www.prophetsecurity.ai/blog/state-of-ai-in-the-soc-2026): Findings from Prophet Security's second annual survey of 250 security leaders and practitioners. 40% run AI in their SOC and 56% are evaluating or piloting it; the average organization leaves about 28% of alerts uninvestigated; a thorough investigation averages about 75 minutes; 72% of AI users cut investigation time by 25% or more; and 46% of in-house AI builds were scrapped, replaced or never reached production.
- [State of AI in Security Operations 2026 (full report)](https://resources.prophetsecurity.ai/state-of-ai-in-security-operations): The full 2026 survey report, with alert volumes and investigation timelines by segment, build-versus-buy results, validation and autonomy practice, hunting frequency and methodology. Gated download.
- [Q3 2026 Prophet Security Quarterly Threat Report](https://resources.prophetsecurity.ai/quarterly-threat-report): Findings from every alert in every customer environment, investigated in full from May through July 2026, covering session hijacking, infostealers, targeted phishing and unmanaged assets. Gated download.
- [AI SOC Statistics: Adoption, Accuracy, and ROI Data](https://www.prophetsecurity.ai/blog/ai-soc-statistics): AI SOC statistics on alert volume, breakout times, adoption, barriers and ROI, each with a dated, citable source.
- [Gartner: The Roles Required for the AI-Enabled Security Operations Center](https://resources.prophetsecurity.ai/gartner-report): Gartner research on staffing the AI-enabled SOC for 2028 and 2030 and the new roles teams need to get value from AI. Gated download.
- [SACR AI SOC Market Landscape 2025](https://resources.prophetsecurity.ai/l/1127683/2026-06-29/j6n7xx/1127683/1782746196hgeHza0v/AI_SOC_Report_Prophet_Security_SACR2025__2_.pdf): Independent market landscape of the AI SOC category from Software Analyst Cyber Research (SACR), covering the capabilities that define success. PDF.
- [AI SOC Agents in the Gartner Hype Cycle for Security Operations](https://www.prophetsecurity.ai/blog/ai-soc-agents-in-gartner-hype-cycle-for-security-operations): Why Gartner added AI SOC agents to the 2025 Hype Cycle for Security Operations, the problems they address and Gartner's recommendations.
- [The Definitive Guide to AI SOC Agents (eBook)](https://resources.prophetsecurity.ai/the-definitive-guide-to-ai-soc-agents): What Prophet Security has learned deploying AI SOC agents in production, how the agents work and how to build a security operation around them. Gated download.
- [The Definitive Guide to Detection Engineering (eBook)](https://resources.prophetsecurity.ai/definitive-guide-detection-engineering): Where detection engineering falls short, why rule-count dashboards can show blind spots as healthy, how to choose which detections to build next, and what an AI-enabled SOC changes. Gated download.
- [The Queue Is the Breach: Your Biggest Risk Is the SOC Queue (eBook)](https://resources.prophetsecurity.ai/ebook-the-queue-is-the-breach): Applies queueing theory to SOC operations to show why headcount and legacy automation alone cannot close the investigation gap. Gated download.
- [The Hype-Free CISO's Guide to Testing an AI SOC Solution](https://resources.prophetsecurity.ai/the-hype-free-cisos-guide-to-testing-an-ai-soc-solution): A vendor-agnostic framework for evaluating AI SOC solutions, built by independent analysts. Gated download.
- [Resources hub](https://www.prophetsecurity.ai/resources): Reports, eBooks, case studies, videos, webinars and events.
## Core Guides: AI SOC and Agentic SOC
- [What Is Agentic SOC? The Complete Guide to How It Works and How to Build One](https://www.prophetsecurity.ai/blog/what-is-agentic-soc): An agentic SOC puts AI agents to work across SOC functions, with people directing and checking the work. How it works, how roles shift and how teams build one.
- [What Is an AI SOC? The Key to Solving Persistent SOC Challenges](https://www.prophetsecurity.ai/blog/ai-soc-key-to-solving-persistent-soc-challenges): What an AI SOC is and how AI agents change alert triage, investigation speed and the analyst's role.
- [What Is an AI SOC Analyst? A Practitioner's Guide for 2026](https://www.prophetsecurity.ai/blog/ai-soc-analyst-a-comprehensive-guide): How an AI SOC analyst investigates every alert at senior-analyst depth, how the mechanics work, what to look for and where the capability stops.
- [What Are AI SOC Agents and How Do They Work?](https://www.prophetsecurity.ai/blog/what-are-ai-soc-agents-how-do-they-work): What AI SOC agents are, how they work, and how to evaluate and adopt them.
- [What Makes an AI SOC Solution Agentic?](https://www.prophetsecurity.ai/blog/what-makes-an-ai-soc-solution-agentic): Four traits that separate agentic AI SOC solutions, which investigate alerts autonomously and adapt in real time, from earlier automation tools.
- [What Is an AI Detection Engineer? A Practitioner's Guide for 2026](https://www.prophetsecurity.ai/blog/what-is-an-ai-detection-engineer): What an AI detection engineer is, the five kinds of work it does, and what changes for the team accountable for detection coverage.
- [What Is AI Threat Hunting and How Does It Work?](https://www.prophetsecurity.ai/blog/threat-hunting-with-ai): What an AI threat hunter does, which hunts run and why, and what changes for a team that hunts continuously.
- [What Is Agentic Security?](https://www.prophetsecurity.ai/blog/what-is-agentic-security-everything-you-should-know): How AI agents investigate and make decisions the way a human analyst would, faster and at scale, and what that means for security operations.
- [What Is an Autonomous SOC, and Can You Build One Today?](https://www.prophetsecurity.ai/blog/what-is-an-autonomous-soc-can-you-build-one-today): What AI can automate in the SOC today, what still needs humans, and how to evaluate autonomous SOC tools.
- [AI SOC Architecture: Integrating with SIEM, SOAR, Case Management and More](https://www.prophetsecurity.ai/blog/ai-soc-architecture-integrating-with-siem-soar-case-management-and-more): How an agentic AI SOC fits into an existing stack of SIEM, SOAR, ITSM and native detection tools.
- [The Human-AI SOC: A Practical Guide to Hybrid Workflows](https://www.prophetsecurity.ai/blog/the-human-ai-soc-a-practical-guide-to-hybrid-workflows): How the Tier 1 role shifts toward reviewing investigations once AI handles rote data collection.
- [How to Build Trust in an AI SOC: A Practical Framework](https://www.prophetsecurity.ai/blog/how-to-build-trust-in-an-ai-soc-a-practical-framework): A framework for transparency, control, coverage, performance metrics and governance.
- [Why Explainability of AI SOC Analyst Platforms Is Important](https://www.prophetsecurity.ai/blog/why-explainability-of-ai-soc-analyst-platforms-is-important): What an explainable AI SOC analyst records: the data reviewed, the questions asked and the evidence behind each conclusion.
- [Will AI Replace the SOC Team? What Gartner Says About AI-Enabled SOC Roles](https://www.prophetsecurity.ai/blog/will-ai-replace-soc-teams): Gartner says the SOC will not shrink by 2030. How AI-enabled roles shift for analysts, engineers and exposure work.
- [Why Human Expertise Remains Irreplaceable in AI-Powered Security Operations](https://www.prophetsecurity.ai/blog/why-human-expertise-remains-irreplaceable-in-ai-powered-security-operations): Where human expertise stays essential in an AI SOC: deciding which questions matter, encoding investigation logic for emerging threats and turning analyst knowledge into detections.
## Buyer's Guides and Evaluation
- [How to Evaluate an Agentic SOC Platform](https://www.prophetsecurity.ai/blog/how-to-evaluate-agentic-soc-platform): Evaluation criteria for investigation depth, response, threat hunting, detection engineering, transparency and privacy.
- [How to Run a Proof of Value (POV) for AI SOC Analysts](https://www.prophetsecurity.ai/blog/how-to-run-a-pov-for-ai-soc-analysts): A step-by-step guide to designing a POV for the AI SOC analyst solutions under evaluation.
- [AI SOC Pricing Models Compared: Per-Alert, Per-Endpoint, Per-Data, and Flat](https://www.prophetsecurity.ai/blog/ai-soc-pricing-models): How each AI SOC pricing model behaves, what inflates the bill and what to compare.
- [How to Build a Business Case for AI in the SOC](https://