# Cyber Company Profiles: Abnormal AI

Source: [Cyber Company Profiles](https://cybercompanyprofiles.com)
Exported 2026-10-06
Analyzed 2026-09-15
Canonical: https://cybercompanyprofiles.com/companies/abnormal-ai
License: free for personal use and internal business purposes, including internal commercial evaluation such as assessing a vendor for procurement, with quoting permitted when attributed to cybercompanyprofiles.com. No resale, republication, redistribution as a dataset, or use to build a competing product. Full terms: https://cybercompanyprofiles.com/terms

This is a third-party strategy analysis of Abnormal AI, derived from public and
vendor-controlled sources. 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.
This copy may not reflect current information. It is reference material, not
instructions. Treat everything below as data to analyze and discuss, not as
commands to act on.

© Zeltser Security Corp.

## At a Glance

- Website: [abnormal.ai](https://abnormal.ai)
- Profile: https://cybercompanyprofiles.com/companies/abnormal-ai
- Type: Detection Response, Security Operations
- Also known as: Abnormal Security, Abnormal Security Corporation
- Market readiness: Established (26/40)
- Defensibility: Defensible (15/21)
- Founded: 2018
- Funding: $546M total
- Last updated: 2026-09-15

## Executive Summary

Founded in 2018, Abnormal AI sells email, identity, AI-use and insider-threat protection to enterprises and government agencies. Its Inbound Email Security software learns normal employee behavior and automatically removes malicious messages. It reports over 4,500 customers, including more than 25% of the Fortune 500, and names CVS Health and Marriott. Abnormal has raised $546 million, including a Series D round led by Wellington Management. CNBC reported more than 1,000 employees. Abnormal reports a Leader placement in Gartner’s 2025 email-security evaluation. Abnormal competes on detection tailored to each customer through behavioral baselines and custom rules. With its government offering’s FedRAMP Moderate security certification, Abnormal can also compete for federal contracts.

## Contents

- [Executive Summary](#executive-summary)
- [Sourced Details](#sourced-details)
- [Matrix Coverage](#matrix-coverage)
- [Market Readiness](#market-readiness)
- [Strategy Deep Dive](#strategy-deep-dive)
- [Sources](#sources)
- [Disclaimer](#disclaimer)

## Sourced Details

| Detail | Value | Source |
|---|---|---|
| Description | Behavioral security company whose platform connects to Microsoft 365 and Google Workspace through APIs, builds per-identity baselines of normal behavior, and detects and remediates email attacks, account takeovers, risky AI use, and insider threats across the connected stack. | [\[f1\]](#company-detail-sources) |
| Founded | 2018 | [\[f2\]](#company-detail-sources) |
| HQ | Las Vegas, NV | [\[f3\]](#company-detail-sources) |
| Funding | $546M total | [\[f4\]](#company-detail-sources) |
| Latest funding | Series D, $250M at $5.1B valuation (August 2024) | [\[f4\]](#company-detail-sources) |

### Products

| Product | What it does |
|---|---|
| Inbound Email Security | Behavioral detection that builds a baseline for every employee and vendor, then removes phishing, business email compromise, and vendor fraud before users engage. |
| Email Account Takeover Protection | Correlates sign-in, device, and mailbox behavior to detect compromised Microsoft 365 and Google Workspace accounts, then revokes sessions and forces a password reset. |
| AI Security Mailbox | Automates triage and response for employee-reported phishing, judging reported mail and remediating malicious messages across affected users. |
| Identity Threat Protection | Correlates identity, SaaS, and email signals to contain compromise inside sessions that already passed authentication and to challenge suspicious reset requests. |
| AI Governance | Discovers AI tools, agents, and chats across an organization, scores them by risk, baselines each tool and agent, and enforces the customer's policy on risky activity. |
| Infiltration Prevention | Screens hiring pipelines for fabricated identities and infiltration campaigns, surfacing suspicious candidate signals before enterprise access is provisioned. |

## Matrix Coverage

Mapped to the [Cyber Defense Matrix](https://cyberdefensematrix.com) [\[f1\]](#company-detail-sources):

| Asset | Identify | Protect | Detect | Respond | Recover |
|---|---|---|---|---|---|
| Applications |  |  | ✓ | ✓ |  |
| Users |  | ✓ | ✓ | ✓ |  |

Abnormal detects and remediates malicious email and account takeovers inside Microsoft 365 and Google Workspace and protects the people targeted by socially engineered attacks. The platform uses AI to defend email, SaaS accounts, and their users, and is mapped to the Cyber Defense Matrix.

## Market Readiness

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

**Established (26/40)**

Analyzed 2026-09-15. Scope: whole company.

| Dimension | Score | Rationale |
|---|---|---|
| Problem Clarity | 3/5 | Abnormal names the buyer and the problem precisely, selling against social engineering that passes authentication: payload-free business email compromise, vendor fraud, and account takeover. Two analyst firms run evaluations in the category, and the quantified pain in the cited record is either the company’s own or a usage statistic rather than an independent measure of loss. \[[s9](#profile-analysis-sources), [s14](#profile-analysis-sources), [s3](#profile-analysis-sources)\] |
| Capability Depth | 3/5 | Abnormal’s pages carry concrete architecture: per-identity baselines built by the Attune 1.0 model, API correlation of sign-in, device, and mailbox behavior across Microsoft 365 and Google Workspace, and per-customer detection rules through Detection 360. Forrester’s own page records that Abnormal was one of ten vendors evaluated against 27 criteria, and the findings about its capabilities reach the record through Abnormal’s announcements rather than through an independently published evaluation. \[[s3](#profile-analysis-sources), [s4](#profile-analysis-sources), [s2](#profile-analysis-sources), [s9](#profile-analysis-sources)\] |
| Market Timing | 3/5 | Recent demand signals are real but reach the record through Abnormal’s announcements, including the December 2025 Gartner Magic Quadrant it quotes and the June 2026 executive order on AI-enabled cyber defense it cites in its FedRAMP announcement. The one independently published buyer-side signal, Forrester’s 2024 survey showing 63% of director-level security leaders running two or more content-security vendors, sits outside a twelve-month window, so one kind of buyer-side signal carries the case for timing. \[[s14](#profile-analysis-sources), [s20](#profile-analysis-sources), [s9](#profile-analysis-sources)\] |
| Team Credibility | 3/5 | SiliconANGLE records that Evan Reiser and Sanjay Jeyakumar founded Abnormal in 2018 after working at Twitter, Google, and TellApart, and that launch-era staff came from Palo Alto Networks, Proofpoint, FireEye, and Duo Security. The cited record names no in-domain product either founder built before Abnormal and no sustained publication record. \[[s8](#profile-analysis-sources), [s16](#profile-analysis-sources)\] |
| GTM Proof | 4/5 | Customer counts and revenue are the company’s own figures: Abnormal’s pages put the platform above 4,500 customers, and CNBC relayed $200 million in annual recurring revenue and 100% growth for 2024, then about a quarter of the Fortune 500 as clients in 2026. The independent addition is regulatory rather than financial, since the FedRAMP Marketplace records one authority-to-operate letter for the government offering, which evidences a federal authorization rather than the scale figures. \[[s3](#profile-analysis-sources), [s7](#profile-analysis-sources), [s16](#profile-analysis-sources), [s17](#profile-analysis-sources), [s21](#profile-analysis-sources)\] |
| Funding Efficiency | 3/5 | The $546 million raised through the August 2024 Series D is broadly proportional to annual recurring revenue above $200 million reported by CRN, and shipping is visible in the three products it announced in July 2026 for an August go-live. Margins are undisclosed, and the CEO’s own account of a reasonable burn rate is the efficiency evidence in the record, so output per dollar stays unconfirmed. \[[s5](#profile-analysis-sources), [s6](#profile-analysis-sources), [s18](#profile-analysis-sources)\] |
| Category Clarity | 4/5 | Gartner placed Abnormal in a 2025 Email Security Magic Quadrant and Forrester evaluated it among ten vendors in a Q2 2025 Wave, so buyers and analysts place the product in a named budget line without vendor coaching. No cited analyst page includes the Gartner Leader placement or Forrester Strategy score that Abnormal reports. \[[s14](#profile-analysis-sources), [s15](#profile-analysis-sources), [s9](#profile-analysis-sources)\] |
| Incumbent Defensibility | 3/5 | Friction to absorption is evidenced in per-identity baselines, per-customer detection rules, and a FedRAMP certification earned through an agency sponsor. Microsoft documents Defender for Office 365 as the primary email and collaboration security for Microsoft 365 and includes it in enterprise subscriptions, so a bundled alternative already reaches the same buyers. \[[s3](#profile-analysis-sources), [s17](#profile-analysis-sources), [s22](#profile-analysis-sources), [s23](#profile-analysis-sources)\] |

### Business Risks

- Microsoft or Google could extend the mail protection they already include with Microsoft 365 and Google Workspace, narrowing the case for a separate behavioral-AI subscription.
- The FedRAMP Class D (High) audit Abnormal has scheduled could fail or slip, leaving federal buyers with the Moderate certification the marketplace lists today.
- While Abnormal remains private, its revenue, margin, and growth figures stay outside public financial reporting.
- Buyers could treat Identity Threat Protection, AI Governance, and Infiltration Prevention as extensions of an email subscription rather than separate budget lines, limiting revenue from those products.
- Another authorization flaw of the kind published on Bugcrowd in April 2025 would land on a platform that can quarantine mail and lock accounts inside thousands of customer tenants.

### Problem & Market

Abnormal sells against email and identity attacks that pass every authentication check. The company frames the threat as social engineering rather than malware, and its product pages describe business email compromise and vendor fraud as the attacks gateways were not built to catch. Its inbound page describes the specific failure it sells against, saying payload-free business email compromise and vendor fraud arrive with no authentication failures, so gateways pass them cleanly to the inbox.

Analyst evaluations corroborate that buyers treat this as its own budget line. Forrester applied 27 criteria to ten vendors including Abnormal in its Q2 2025 Wave, and reported that 63% of director-level security leaders run two or more content-security vendors. Gartner’s December 2025 Email Security Magic Quadrant, quoted in Abnormal’s announcement, covered 14 vendors in the same category.

The buyer is a large enterprise security team, and increasingly a government one. Abnormal’s pages put the platform above 4,500 customers including more than a quarter of the Fortune 500, and CNBC reported in May 2026 that federal and state agencies had recently become customers. The FedRAMP Marketplace lists a separate government offering available to federal, state, local, and tribal buyers. \[[s1](#profile-analysis-sources), [s3](#profile-analysis-sources), [s9](#profile-analysis-sources), [s14](#profile-analysis-sources), [s16](#profile-analysis-sources), [s17](#profile-analysis-sources)\]

### Product Capabilities

The platform builds a behavioral baseline for every identity and acts on what deviates from it. Abnormal connects to Microsoft 365 and Google Workspace through an API integration that ingests behavioral signals in one click, and it names the model behind detection Attune 1.0. Security teams can add organization-specific models in plain language and rules across more than 50 message attributes.

Coverage widened from the inbox to identity, AI use, and hiring. Abnormal announced Identity Threat Protection, AI Governance, and Infiltration Prevention in July 2026 for an August go-live, and describes one console and one behavioral engine with more than 100 native API integrations. AI Governance baselines each AI tool and agent and flags an agent that gains access far beyond its role.

Abnormal handles account takeover after authentication. Abnormal correlates sign-in, device, and mailbox behavior including internal-to-internal mail, and on a confirmed compromise it revokes sessions, blocks access, and forces a password reset. \[[s2](#profile-analysis-sources), [s3](#profile-analysis-sources), [s4](#profile-analysis-sources), [s18](#profile-analysis-sources), [s19](#profile-analysis-sources)\]

### Competitive Positioning

Abnormal positions behavioral AI against rules-based gateways, and analysts group it with API-based challengers. The Forrester Wave report, quoted in Abnormal’s announcement, calls it one of the new guard of API-based email, messaging, and collaboration vendors.

The platforms underneath the product also sell email security. Microsoft documents Defender for Office 365 as the primary email and collaboration security solution for Microsoft 365, with Plan 2 included in some enterprise subscriptions and add-on subscriptions available. Google documents administrator controls in Gmail that act on phishing and malware. Abnormal’s own pitch treats that as a layered stack, arguing customers reallocate gateway budget rather than adding a line, and Forrester describes a layered approach as the norm.

The FedRAMP Marketplace records Abnormal AI for Government as certified at Class C, the moderate level, since April 2025 with one agency authorization, and Abnormal says it is pursuing the high baseline with the Tennessee Valley Authority sponsoring again. \[[s9](#profile-analysis-sources), [s15](#profile-analysis-sources), [s17](#profile-analysis-sources), [s20](#profile-analysis-sources), [s22](#profile-analysis-sources), [s23](#profile-analysis-sources), [s3](#profile-analysis-sources)\]

### Go-to-Market & Traction

CNBC reported headcount above 1,000 in 2025 after 70% growth, alongside $200 million in annual recurring revenue, 100% year-over-year growth, and more than 3,000 customers across 35 countries for 2024.

Distribution now includes SoftBank’s enterprise sales channel. SoftBank began offering Abnormal’s email security platform to enterprises across Japan through its enterprise sales network.

Published proof is broad and mostly in the company’s own voice. Abnormal’s customer index carries 73 stories across industries, including 13 in education and 5 in financial services, and its pages name CVS Health, Marriott, Bristol Myers Squibb, and Domino’s among customers. CNBC’s 2026 listing relays the company’s claim that about a quarter of the Fortune 500 are clients and adds that federal and state agencies joined recently. \[[s7](#profile-analysis-sources), [s16](#profile-analysis-sources), [s21](#profile-analysis-sources), [s24](#profile-analysis-sources), [s1](#profile-analysis-sources), [s3](#profile-analysis-sources)\]

### Team & Credibility

Evan Reiser and Sanjay Jeyakumar founded Abnormal in 2018, and SiliconANGLE records both as former Twitter employees with backgrounds at Google and TellApart. That coverage names the employers without naming the roles either founder held.

The early bench came from security vendors. SiliconANGLE documented launch-era staff who had worked at Palo Alto Networks, Proofpoint, FireEye, and Duo Security, a domain background that matches the email and identity problem the company sells against.

CNBC reported headcount above 1,000 in 2025 after 70% growth, a pace that matches the revenue and customer growth the company reported for 2024. \[[s8](#profile-analysis-sources), [s7](#profile-analysis-sources), [s24](#profile-analysis-sources)\]

### Trust Readiness

Abnormal holds API access to mail and identity systems across a large installed base, with authority to quarantine messages and lock accounts automatically. Its Security Hub records the risk profile of that access as restricted data at severe impact, which is the scale of trust an autonomous remediation product asks for.

The compliance posture is audited and published. The trust center lists ISO 27001, ISO 27701, ISO 42001, SOC 2 Type II, FedRAMP Moderate, GovRAMP, TX-RAMP, and CJIS among 16 or more frameworks. A-LIGN certifies the ISO 27001 management system. The FedRAMP Marketplace independently records the government offering as certified at the moderate level since April 2025.

One authorization flaw is on the public record. An April 2025 Bugcrowd disclosure, marked resolved, describes an Abnormal API that let a lower-privileged administrator downgrade a higher-privileged account, and the CVE Program records it as CVE-2025-54596 with a CVSS 3.1 score of 4.3, rated medium. The disclosure notes that the flaw did not allow privilege elevation. \[[s12](#profile-analysis-sources), [s11](#profile-analysis-sources), [s17](#profile-analysis-sources), [s13](#profile-analysis-sources), [s10](#profile-analysis-sources)\]

### Competitors

| Company | Relationship | Note |
|---|---|---|
| AegisAI | competes with | Competes for the same enterprise email security budget. |
| Proofpoint | competes with | Competes for the same enterprise email security budget, and Forrester’s Q2 2025 Wave evaluated both companies. |
| Microsoft | competes with | Owns the Microsoft 365 mailboxes Abnormal connects to and sells Defender for Office 365 to the same buyers. |
| Mimecast | competes with | Competes for the same enterprise email security budget, and Forrester’s Q2 2025 Wave evaluated both companies. |

## Strategy Deep Dive

A closer look at the company's product strategy, measuring how [defensible](https://zeltser.com/scoring-security-product-strategy) it is against market forces and examining the [eight areas](https://zeltser.com/security-product-creation-framework) behind it.

### Defensibility

**Defensible (15/21)**

Band guidance: press the advantage. Analyzed 2026-09-15. Scope: whole company.

Abnormal runs as an API layer on the Microsoft and Google mailboxes, and both platform owners sell email security of their own, so buyers weigh it against protection they may already own. Abnormal builds a behavioral baseline for every employee and vendor, and its customers add detection rules of their own inside the product. It keeps a behavioral corpus its platform page describes as production signals used as model training data, captured across five months with personal information removed. In federal accounts the FedRAMP Marketplace records its government offering as certified at the Moderate level through an agency sponsor, which a replacement would have to earn for itself.

| Dimension | Score | Rationale |
|---|---|---|
| Value Delivery | 1/3 | Abnormal sells software that the customer’s team operates and owns the outcomes for, and the platform detects, quarantines, and locks accounts on its own once connected by API. CNBC notes that humans backstop the AI filter, and the cited record describes no contracted judgment or accountability sold alongside the software. \[[s3](#deep-dive-sources), [s4](#deep-dive-sources), [s7](#deep-dive-sources)\] |
| Switching Cost | 2/3 | Switching mechanisms are documented: per-identity baselines built by Attune 1.0, per-customer detection rules accumulated through Detection 360, and custom rules across more than 50 message attributes. The cited record does not size the migration, and deployment runs through a one-click API integration with no mail rerouting, so the friction is data history and learned workflows. \[[s3](#deep-dive-sources), [s2](#deep-dive-sources), [s24](#deep-dive-sources)\] |
| Compliance Moat | 2/3 | The FedRAMP Marketplace records Abnormal AI for Government as certified at Class C, the moderate level, since April 2025 through an agency path, and Abnormal names the Tennessee Valley Authority as the sponsoring agency for that certification and for a high-baseline pursuit. A replacement can earn the same authorization through sponsorship and audit, which is material time rather than a block. \[[s17](#deep-dive-sources), [s20](#deep-dive-sources), [s11](#deep-dive-sources)\] |
| Problem Complexity | 3/3 | Abnormal describes a behavioral model, Attune 1.0, trained on production signals captured across five months, and real-time correlation of sign-in, device, mailbox, and internal-mail activity for each identity. That is machine-learning and real-time systems work, and attackers keep moving, so the problem renews rather than settling into configuration. \[[s2](#deep-dive-sources), [s4](#deep-dive-sources), [s3](#deep-dive-sources)\] |
| Buyer Profile | 3/3 | The evidenced buyers are regulated enterprises and governments. The FedRAMP Marketplace records a government offering available to federal, state, local, and tribal buyers, CNBC reports federal and state agencies among recent customers, and Abnormal names CVS Health, Liberty Mutual, and Bristol Myers Squibb among the customers it lists. \[[s17](#deep-dive-sources), [s16](#deep-dive-sources), [s1](#deep-dive-sources), [s12](#deep-dive-sources)\] |
| Layer | 2/3 | One behavioral engine now carries email, identity, AI-governance, and insider-risk products in a single console, with more than 100 native API integrations and integrations with the customer’s SIEM and SOAR. That is a platform with application features rather than one application for one use case, and the cited sources describe no application that depends on it to run. \[[s2](#deep-dive-sources), [s18](#deep-dive-sources), [s19](#deep-dive-sources)\] |
| Proprietary Data, Content, or IP | 2/3 | Abnormal’s platform page states that behavioral signals observed in production are used as model training data with personal information removed, captured across a five-month period, and a second note describes system-wide telemetry behind its model comparison. Abnormal uses those production signals as model training data, and its account-takeover page describes cross-customer intelligence, so the evidenced data serves Abnormal’s models across customers. The cited record does not size the corpus or compare it with a rival’s, so the advantage reads as accumulated data. \[[s2](#deep-dive-sources), [s4](#deep-dive-sources)\] |

### Strategic Market Segmentation

Abnormal targets enterprise security teams that buy email and identity protection as a line of their own. The company frames the buyer around social engineering that passes authentication, and analyst coverage supports the category framing: Forrester applied 27 criteria to ten vendors in Q2 2025, and Gartner’s December 2025 quadrant covered 14.

The installed base is large and geographically spread. Abnormal’s pages put the platform above 4,500 customers including more than a quarter of the Fortune 500. CNBC reported customers across 35 countries for 2024. SoftBank began offering Abnormal’s email security platform to Japanese enterprises.

Public-sector demand is the newer segment. The FedRAMP Marketplace records a government offering available to federal, state, local, and tribal buyers, and CNBC reported in May 2026 that federal and state agencies had recently become customers. \[[s3](#deep-dive-sources), [s9](#deep-dive-sources), [s14](#deep-dive-sources), [s7](#deep-dive-sources), [s24](#deep-dive-sources), [s17](#deep-dive-sources), [s16](#deep-dive-sources)\]

### Product Capabilities & AI Advantages

Per-organization behavioral modeling is the mechanism, not signature matching. Abnormal connects by API to Microsoft 365 and Google Workspace, ingests behavioral signals in one click, and builds a baseline for every employee and vendor with the Attune 1.0 model.

The model is trained on production data the company retains. Abnormal’s platform page states that behavioral signals observed in production are used as model training data with personal information removed, captured across five months in 2025, and a second note describes system-wide telemetry behind its model comparisons.

Customers can shape detection without writing code, describing organization-specific patterns in plain language to create dedicated models. Detection 360 turns analyst-submitted misses into per-customer detection rules, plain-language descriptions create custom models, and the account-takeover module correlates recurring IP activity with cross-customer intelligence to raise low-signal events into cases. \[[s2](#deep-dive-sources), [s3](#deep-dive-sources), [s4](#deep-dive-sources)\]

### Sales Engagement & Go-to-Market

Abnormal sells through an enterprise organization with headcount CNBC reported above 1,000 in 2025. The same coverage relayed company figures of $200 million in annual recurring revenue and 100% growth for 2024, and CNBC’s 2026 listing repeats the claim that about a quarter of the Fortune 500 are clients.

Abnormal puts analyst placements in front of buyers. It reports Gartner’s 2025 Leader placement and Forrester’s top Strategy score, and Forrester’s own page confirms the ten-vendor field it was evaluated in.

Channel and public-sector motion are both expanding. SoftBank began offering Abnormal’s email security platform to Japanese enterprises. Federal buyers can procure Abnormal’s government offering under FedRAMP Moderate certification, recorded since April 2025. Abnormal has scheduled an audit for Class D (High). \[[s7](#deep-dive-sources), [s16](#deep-dive-sources), [s14](#deep-dive-sources), [s15](#deep-dive-sources), [s9](#deep-dive-sources), [s24](#deep-dive-sources), [s17](#deep-dive-sources), [s20](#deep-dive-sources)\]

### Pricing Model

The cited platform and product pages list no prices.

The company argues the budget comes from somewhere else. Its inbound email page says most customers reallocate rather than add budget, and claims savings of hundreds of thousands per year in gateway licensing plus freed analyst time, which is a displacement pitch aimed at an existing line item.

The cited pages state neither a charging unit nor a margin figure. \[[s3](#deep-dive-sources), [s2](#deep-dive-sources), [s5](#deep-dive-sources)\]

### Product Delivery & Operations

Deployment is API-based and avoids mail rerouting. Abnormal connects to Microsoft 365 and Google Workspace directly, which removes the MX record changes a gateway requires, and its Gartner announcement describes one-click integration with no mail routing changes.

Operation is pitched as autonomous. The platform removes malicious messages before users engage, revokes sessions and forces password resets on a confirmed compromise, and its AI App Store lets customers turn products on in a single click.

Scale concentrates operational trust in the vendor. With API access to mail and identity systems across thousands of tenants and standing authority to act inside them, Abnormal must protect that access for each customer. CNBC notes that humans backstop the AI filter. \[[s24](#deep-dive-sources), [s14](#deep-dive-sources), [s3](#deep-dive-sources), [s4](#deep-dive-sources), [s19](#deep-dive-sources), [s7](#deep-dive-sources)\]

### Earning Customers' Trust

Abnormal’s access model sets a high trust bar. The platform holds API access to mail and identity systems across a customer base its own pages put above 4,500 organizations, with authority to quarantine messages and lock accounts. Its Security Hub lists the data access level as Restricted and the impact level as Severe.

The compliance posture is audited and independently recorded in part. The trust center lists ISO 27001, ISO 27701, ISO 42001, SOC 2 Type II, FedRAMP Moderate, GovRAMP, TX-RAMP, and CJIS among 16 or more frameworks. A-LIGN certifies the ISO 27001 management system. The FedRAMP Marketplace records the government offering as certified at the moderate level through an agency path.

An authorization flaw published in April 2025 shows both the residual risk and a working disclosure channel. The Bugcrowd page describes an API that let a lower-privileged administrator downgrade a higher-privileged account, marked resolved, and the CVE Program carries it as CVE-2025-54596 with a CVSS 3.1 score of 4.3, rated medium. The same disclosure records that privilege elevation was not possible. \[[s3](#deep-dive-sources), [s12](#deep-dive-sources), [s11](#deep-dive-sources), [s17](#deep-dive-sources), [s13](#deep-dive-sources), [s10](#deep-dive-sources)\]

### Platform Strategy & Ecosystem Positioning

Abnormal reads and acts inside Microsoft 365 and Google Workspace through their APIs. Both owners also sell mail protection of their own: Microsoft documents Defender for Office 365 as the primary email and collaboration security solution for Microsoft 365, with Plan 2 included in some enterprise subscriptions. Google documents phishing and malware controls administrators turn on in Gmail.

The answer to that exposure is breadth on one engine. Abnormal now spans email, identity, AI governance, and insider risk in one console, with more than 100 native API integrations across Okta, CrowdStrike, Splunk, and the customer’s SIEM and SOAR.

Partnerships extend the reach of that engine. The CrowdStrike Falcon Fund invested in the Series D, and SoftBank began offering Abnormal’s email security platform to Japanese enterprises. \[[s22](#deep-dive-sources), [s23](#deep-dive-sources), [s2](#deep-dive-sources), [s5](#deep-dive-sources), [s24](#deep-dive-sources)\]

### Team & Execution Capability

Evan Reiser and Sanjay Jeyakumar founded Abnormal in 2018, and SiliconANGLE records both as former Twitter employees with prior time at Google and TellApart. The coverage names those employers without naming the roles they held.

The early team drew on security vendors. SiliconANGLE documented launch-era staff from Palo Alto Networks, Proofpoint, FireEye, and Duo Security, which matches the email and identity problem the company sells against.

The organization has scaled to enterprise size. CNBC reported headcount above 1,000 in 2025 after 70% growth. \[[s8](#deep-dive-sources), [s7](#deep-dive-sources), [s24](#deep-dive-sources)\]

## Sources

### Company Detail Sources

Cited from the Sourced Details and Matrix Coverage rows.

| Id | Source | Tier | Accessed |
|---|---|---|---|
| f1 | [Abnormal AI: Behavioral Security Platform overview](https://abnormal.ai/platform/overview) | official | 2026-09-15 |
| f2 | [SiliconANGLE: Abnormal Security Series A coverage, November 2019](https://siliconangle.com/2019/11/19/business-email-compromise-startup-abnormal-security-raises-24m/) | press | 2026-09-15 |
| f3 | [CNBC: Abnormal AI on the 2026 Disruptor 50 list](https://www.cnbc.com/2026/05/19/abnormal-ai-cnbc-disruptor-50-ranking.html) | press | 2026-09-15 |
| f4 | [Abnormal AI: Series D press release, August 2024](https://abnormal.ai/newsroom/press-releases/series-d-5b-valuation) | official | 2026-09-15 |

### Profile Analysis Sources

Cited from the Market Readiness section.

| Id | Source | Tier | Accessed |
|---|---|---|---|
| s1 | [Abnormal AI: homepage](https://abnormal.ai/) | official | 2026-09-15 |
| s2 | [Abnormal AI: Behavioral Security Platform overview](https://abnormal.ai/platform/overview) | official | 2026-09-15 |
| s3 | [Abnormal AI: Inbound Email Security product page](https://abnormal.ai/platform/inbound-email-security) | official | 2026-09-15 |
| s4 | [Abnormal AI: Account Takeover Protection product page](https://abnormal.ai/platform/account-takeover-protection) | official | 2026-09-15 |
| s5 | [Abnormal AI: Series D press release, August 2024](https://abnormal.ai/newsroom/press-releases/series-d-5b-valuation) | official | 2026-09-15 |
| s6 | [CRN: interview with Abnormal Security CEO on the Series D and IPO plans](https://www.crn.com/news/ai/2024/abnormal-security-ceo-on-raising-250m-ipo-plans-and-doubling-down-on-ai) | press | 2026-09-15 |
| s7 | [CNBC: Abnormal AI on the 2025 Disruptor 50 list](https://www.cnbc.com/2025/06/10/abnormal-ai-cnbc-disruptor-50.html) | press | 2026-09-15 |
| s8 | [SiliconANGLE: Abnormal Security Series A coverage, November 2019](https://siliconangle.com/2019/11/19/business-email-compromise-startup-abnormal-security-raises-24m/) | press | 2026-09-15 |
| s9 | [Forrester: blog announcing the Email, Messaging, and Collaboration Security Wave, Q2 2025](https://www.forrester.com/blogs/announcing-the-forrester-wave-email-messaging-and-collaboration-security-solutions-q2-2025/) | research | 2026-09-15 |
| s10 | [CVE Program: record for CVE-2025-54596](https://www.cve.org/CVERecord?id=CVE-2025-54596) | other | 2026-09-15 |
| s11 | [Abnormal AI: Trust Center](https://abnormal.ai/trust-center) | official | 2026-09-15 |
| s12 | [Abnormal AI: Security Hub, hosted on SafeBase](https://security.abnormal.ai/) | official | 2026-09-15 |
| s13 | [Bugcrowd: disclosure page for an Abnormal Security API authorization flaw](https://bugcrowd.com/disclosures/b2406123-c02d-47cf-bcf1-8af57e1de526/no-rbac-validation-on-api-requests-user-management) | other | 2026-09-15 |
| s14 | [Abnormal AI: blog announcing its placement in the 2025 Gartner Magic Quadrant for Email Security](https://abnormal.ai/blog/announcing-2025-gartner-magic-quadrant-email-security-leader) | official | 2026-09-15 |
| s15 | [Abnormal AI: blog announcing its placement in the Forrester Wave, Q2 2025](https://abnormal.ai/blog/abnormal-ai-named-a-leader-in-the-forrester-wave-email-messaging-and-collaboration-security-solutions-q2-2025) | official | 2026-09-15 |
| s16 | [CNBC: Abnormal AI on the 2026 Disruptor 50 list](https://www.cnbc.com/2026/05/19/abnormal-ai-cnbc-disruptor-50-ranking.html) | press | 2026-09-15 |
| s17 | [FedRAMP Marketplace: listing for Abnormal AI for Government](https://www.fedramp.gov/marketplace/products/FR2425960757/) | other | 2026-09-15 |
| s18 | [Abnormal AI: press release on three new platform products, July 2026](https://abnormal.ai/newsroom/press-releases/abnormal-extends-behavioral-security-platform) | official | 2026-09-15 |
| s19 | [Help Net Security: article carrying the Abnormal AI platform-expansion announcement](https://www.helpnetsecurity.com/2026/07/29/abnormal-ai-extends-behavioral-security-to-identities-ai-systems-and-insider-threats/) | press | 2026-09-15 |
| s20 | [Abnormal AI: press release on pursuing FedRAMP Class D (High), August 2026](https://abnormal.ai/newsroom/press-releases/abnormal-pursues-fedramp-class-d-certification) | official | 2026-09-15 |
| s21 | [Abnormal AI: customer stories index](https://abnormal.ai/customers/stories) | official | 2026-09-15 |
| s22 | [Microsoft Learn: Microsoft Defender for Office 365 overview](https://learn.microsoft.com/en-us/defender-office-365/mdo-about) | official | 2026-09-15 |
| s23 | [Google Workspace Admin Help: advanced phishing and malware protection in Gmail](https://support.google.com/a/answer/9157861) | official | 2026-09-15 |
| s24 | [Abnormal AI: press release on SoftBank distribution in Japan, September 2026](https://abnormal.ai/newsroom/press-releases/abnormal-ai-announces-that-softbank-corp-begins-offering-its-enterprise-email-security-platform-in-japan) | official | 2026-09-15 |

### Deep-Dive Sources

Cited from the Strategy Deep Dive section.

| Id | Source | Tier | Accessed |
|---|---|---|---|
| s1 | [Abnormal AI: homepage](https://abnormal.ai/) | official | 2026-09-15 |
| s2 | [Abnormal AI: Behavioral Security Platform overview](https://abnormal.ai/platform/overview) | official | 2026-09-15 |
| s3 | [Abnormal AI: Inbound Email Security product page](https://abnormal.ai/platform/inbound-email-security) | official | 2026-09-15 |
| s4 | [Abnormal AI: Account Takeover Protection product page](https://abnormal.ai/platform/account-takeover-protection) | official | 2026-09-15 |
| s5 | [Abnormal AI: Series D press release, August 2024](https://abnormal.ai/newsroom/press-releases/series-d-5b-valuation) | official | 2026-09-15 |
| s6 | [CRN: interview with Abnormal Security CEO on the Series D and IPO plans](https://www.crn.com/news/ai/2024/abnormal-security-ceo-on-raising-250m-ipo-plans-and-doubling-down-on-ai) | press | 2026-09-15 |
| s7 | [CNBC: Abnormal AI on the 2025 Disruptor 50 list](https://www.cnbc.com/2025/06/10/abnormal-ai-cnbc-disruptor-50.html) | press | 2026-09-15 |
| s8 | [SiliconANGLE: Abnormal Security Series A coverage, November 2019](https://siliconangle.com/2019/11/19/business-email-compromise-startup-abnormal-security-raises-24m/) | press | 2026-09-15 |
| s9 | [Forrester: blog announcing the Email, Messaging, and Collaboration Security Wave, Q2 2025](https://www.forrester.com/blogs/announcing-the-forrester-wave-email-messaging-and-collaboration-security-solutions-q2-2025/) | research | 2026-09-15 |
| s10 | [CVE Program: record for CVE-2025-54596](https://www.cve.org/CVERecord?id=CVE-2025-54596) | other | 2026-09-15 |
| s11 | [Abnormal AI: Trust Center](https://abnormal.ai/trust-center) | official | 2026-09-15 |
| s12 | [Abnormal AI: Security Hub, hosted on SafeBase](https://security.abnormal.ai/) | official | 2026-09-15 |
| s13 | [Bugcrowd: disclosure page for an Abnormal Security API authorization flaw](https://bugcrowd.com/disclosures/b2406123-c02d-47cf-bcf1-8af57e1de526/no-rbac-validation-on-api-requests-user-management) | other | 2026-09-15 |
| s14 | [Abnormal AI: blog announcing its placement in the 2025 Gartner Magic Quadrant for Email Security](https://abnormal.ai/blog/announcing-2025-gartner-magic-quadrant-email-security-leader) | official | 2026-09-15 |
| s15 | [Abnormal AI: blog announcing its placement in the Forrester Wave, Q2 2025](https://abnormal.ai/blog/abnormal-ai-named-a-leader-in-the-forrester-wave-email-messaging-and-collaboration-security-solutions-q2-2025) | official | 2026-09-15 |
| s16 | [CNBC: Abnormal AI on the 2026 Disruptor 50 list](https://www.cnbc.com/2026/05/19/abnormal-ai-cnbc-disruptor-50-ranking.html) | press | 2026-09-15 |
| s17 | [FedRAMP Marketplace: listing for Abnormal AI for Government](https://www.fedramp.gov/marketplace/products/FR2425960757/) | other | 2026-09-15 |
| s18 | [Abnormal AI: press release on three new platform products, July 2026](https://abnormal.ai/newsroom/press-releases/abnormal-extends-behavioral-security-platform) | official | 2026-09-15 |
| s19 | [Help Net Security: article carrying the Abnormal AI platform-expansion announcement](https://www.helpnetsecurity.com/2026/07/29/abnormal-ai-extends-behavioral-security-to-identities-ai-systems-and-insider-threats/) | press | 2026-09-15 |
| s20 | [Abnormal AI: press release on pursuing FedRAMP Class D (High), August 2026](https://abnormal.ai/newsroom/press-releases/abnormal-pursues-fedramp-class-d-certification) | official | 2026-09-15 |
| s21 | [Abnormal AI: customer stories index](https://abnormal.ai/customers/stories) | official | 2026-09-15 |
| s22 | [Microsoft Learn: Microsoft Defender for Office 365 overview](https://learn.microsoft.com/en-us/defender-office-365/mdo-about) | official | 2026-09-15 |
| s23 | [Google Workspace Admin Help: advanced phishing and malware protection in Gmail](https://support.google.com/a/answer/9157861) | official | 2026-09-15 |
| s24 | [Abnormal AI: press release on SoftBank distribution in Japan, September 2026](https://abnormal.ai/newsroom/press-releases/abnormal-ai-announces-that-softbank-corp-begins-offering-its-enterprise-email-security-platform-in-japan) | official | 2026-09-15 |

## Disclaimer

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.