Small business▼−4Since September 2026: 6% → 2%, −4 points. Inside the 11-point floor: within noise. Read over the models both editions asked.
2%from 6% in September 2026
Inside the floor by 7 pointsRank 5 → 9 of 56
Abnormal · 3 categories · RankedAbnormal AI
112Judge labels
8First choices
6Negative labels
14 / 14Models named it
3Categories
October 2026 Edition. Every number here is derived from the raw labels under vendor table v2026-10.8, every buyer segment counted.
Best standing
10% in Email security for enterprise buyers
Rank 7 of 76 in the mid-market standingaccepted challenger
2 of 14 models made it the first choice on the direct prompt; 2% of its 42 labels there were negative.
What the models named instead of Abnormal AI →By buyer segmentRead the same way at every buyer size.
Small business10 of 66
2%19 labels
accepted challenger
Mid-market7 of 76
4%42 labels
accepted challenger
Enterprise3 of 61
10%47 labels
accepted challenger
In email security · each standing computed within its segment · bars are 0 to 100 · the accent bar is the product's own best reading
Standing by category
Every category where a model named Abnormal AI for a mid-market B2B company. Share is first choices across the direct, paraphrase, budget and scale prompts; rank is within every product named in that category.
| Category | Function | Share | Rank | Negative rate | Labels | Quadrant | Since September 2026 |
|---|---|---|---|---|---|---|---|
| Email security | Security operations | 4% | 7 of 76 | 2% | 42 | accepted challenger | ▼−10Since September 2026: 12% → 2%, −10 points. Inside the 11-point floor: within noise. Read over the models both editions asked. |
| AI security platforms | Security operations | 0% | 144 of 233 | 0% | 1 | under 10 labels · led by SentinelOne Singularity at 11% | |
| Enterprise browsers | Security operations | 0% | 30 of 80 | 0% | 1 | under 10 labels · led by Google Password Manager at 26% |
Movement
Since September 2026, Abnormal AI held within the floor in all three categories: no change cleared 11 points.
Best category · Email security · by segment
Mid-market▼−10Since September 2026: 12% → 2%, −10 points. Inside the 11-point floor: within noise. Read over the models both editions asked.
2%from 12% in September 2026
Inside the floor by 1 pointRank 2 → 7 of 71
Enterprise▲+2Since September 2026: 10% → 11%, +2 points. Inside the 11-point floor: within noise. Read over the models both editions asked.
11%from 10% in September 2026
Inside the floor by 9 pointsRank 3 of 58, unchanged
All categories · Mid-market
| Category | September 2026 | Now | Change | Reading | Rank |
|---|---|---|---|---|---|
| Email security | 12% | 2% | ▼−10Since September 2026: 12% → 2%, −10 points. Inside the 11-point floor: within noise. Read over the models both editions asked. | Within noise | Rank 2 → 7 of 76 |
| AI security | 0% | 0% | newNew since September 2026: not ranked then, 0% now. | New this edition | Rank 144 of 233, unchanged |
| Enterprise browsers | 0% | 0% | newNew since September 2026: not ranked then, 0% now. | New this edition | Rank 30 of 80, unchanged |
Shares here are read over the models both editions asked, so they can differ by a point or two from the standing above, which counts every model in this edition.
The floor is 11 points of share, measured: how far the models move a leader on their own when the same questions are asked twice with nothing changed. A larger change is movement; a smaller one is noise, and both are shown. Movement is read over the twelve models both editions asked; GPT-6 Luna, Muse Glimmer 30B joined this edition and are in the standing but not yet in the comparison. How the floor is measured · The editions
By model
How each model treated Abnormal AI across every prompt where it was named for a mid-market B2B company. Fourteen models, six prompts per category.ShowHide
By model
How each model treated Abnormal AI across every prompt where it was named for a mid-market B2B company. Fourteen models, six prompts per category.
ShowHide| Model | First choice | Alternative | Mention | Negative | Labels |
|---|---|---|---|---|---|
| Claude Haiku 4.5 | 0 | 0 | 2 | 0 | 2 |
| GPT-5.4 mini | 0 | 2 | 0 | 0 | 2 |
| Gemini 3.5 Flash | 2 | 3 | 0 | 0 | 5 |
| Perplexity Sonar | 0 | 0 | 1 | 0 | 1 |
| Grok 4.20 | 1 | 2 | 2 | 0 | 5 |
| Mistral Small | 0 | 0 | 1 | 0 | 1 |
| DeepSeek V4 Flash | 2 | 2 | 0 | 1 | 5 |
| Llama 4 Maverick | 0 | 1 | 2 | 0 | 3 |
| Qwen 3.7 Flash | 0 | 2 | 0 | 0 | 2 |
| Kimi K2 | 0 | 2 | 2 | 0 | 4 |
| GLM 4.7 FlashX | 1 | 3 | 2 | 0 | 6 |
| MiniMax M2.5 | 0 | 1 | 0 | 0 | 1 |
| GPT-6 Luna | 1 | 2 | 0 | 0 | 3 |
| Muse Glimmer 30B | 1 | 1 | 2 | 0 | 4 |
By framing
Which of the six questions produced the naming. By model says how often; this says asked what. The first-choice count on the right carries the marks of the models that produced it.
FramingLabels by classFirst choices
Direct25 labels4
Paraphrase17 labels2
Comparative31 labels7not counted in share
Budget-constrained9 labelsNone
Scale-constrained15 labels2
Negative15 labels5not counted in share
First choiceAlternativeMentionNegative112 labels in all, every segment counted; 8 of the 20 first choices count toward share, since the comparative and negative framings do not. The bar is one segment per label class, to scale within the framing.
What the models said for it
Verbatim evidence the judge attached to positive labels.
“I'd shortlist Abnormal AI as the best add-on for catching business-email-compromise (BEC), vendor impersonation, and other attacks that slip past standard filtering” GPT-6 Luna · Email security · direct prompt · first choice
“Abnormal Security is the top AI-native option for teams that need API-based detection of social-engineering attacks without gateway overhead.” Muse Glimmer 30B · Email security · comparative prompt · first choice
“These solutions (such as Abnormal Security, Avanan, or Sublime) connect directly to your email provider via native APIs.” Gemini 3.5 Flash · Email security · negative prompt · first choice
“The pioneer of behavioral-AI email security... Unrivaled at stopping BEC and VIP impersonation” Gemini 3.5 Flash · Email security · comparative prompt · first choice
And against it
Verbatim evidence attached to negative labels. A warning on a product with few labels is a warning; on a product with many, it is one voice among them.
“Abnormal Security (caution, not avoid)” DeepSeek V4 Flash · Email security · negative prompt · soft negative
Named alongside
The products named in the same answers as Abnormal AI, over the 111 answers that named it. Took the first choice instead counts the answers where the other product was the first choice and Abnormal AI was named but was not.ShowHide
Named alongside
The products named in the same answers as Abnormal AI, over the 111 answers that named it. Took the first choice instead counts the answers where the other product was the first choice and Abnormal AI was named but was not.
ShowHideProductSame answerTook the first choice insteadHead to head
A head-to-head page exists where both products are among a category's top eight. The other rows are the same fact without a page behind them, so they link to the product instead.
What carried it into the answer
The sites and pages cited by the answers that named Abnormal AI. A fact about retrieval, not a lever on the model.
Citations exist only for the models that return a source list, five of the fourteen in this edition, so these counts come from 106 of the 111 answers that named Abnormal AI and are not a share of its labels.
Domains cited
expertinsights.com52
kaseya.com51
ironscales.com42
proofpoint.com41
sublime.security41
huntress.com39
gartner.com37
comparitech.com34
checkpoint.com33
g2.com31
401 of the 401 domain citations in answers naming Abnormal AI came from somebody else's page.
Pages cited
Pages are listed as the models cited them.
Search and answers
abnormal.ai is not in Google's top ten for any of the category's searches. In the answers, Abnormal AI takes 4% of first choices and Microsoft Defender for Office 365 takes 32%.ShowHide
Search and answers
abnormal.ai is not in Google's top ten for any of the category's searches. In the answers, Abnormal AI takes 4% of first choices and Microsoft Defender for Office 365 takes 32%.
ShowHideIn search
Google, US estimatesPosition for “email security platform”
–
not in the top ten
Position for “email security platforms”
–
not in the top ten
- Searches for its name, Google
- 8,100 a month (“abnormal ai”)
- AI search demand for its name, est.
- 130 a month
- Its company's site
- abnormal.ai shared with Abnormal's other products; the figures are the site's
- Organic visits to its site
- about 30,467 a month
- Searches its site ranks for
- 3,559 · 187 in the top three
- Sites linking to it
- 5,070
- Paid Google search
- 1 searches its ads show for, about 1 clicks and $6 a month (estimate)
In answers
This editionShare of first choices
4%
rank 7 of 76 in email security
Segment leader
32%
Microsoft Defender for Office 365
- First choices
- 8 across its categories
- Named in
- 111 answers
- Its own site cited
- in 106 of the answers that named it
Search figures are US estimates from DataForSEO, read September 28, 2026; AI search demand is its modeled, directional estimate, not a count of queries to any assistant. The answers are this edition's. Two measurements side by side: neither is read as the cause of the other.
Names read as Abnormal AI
What the judge wrote, as written, with how often. The vendor table decides that these count as Abnormal AI; a claim can dispute any of them.
Abnormal Security 26Abnormal 5
Follow Abnormal AI
An email the morning each edition publishes: where this product moved, where it held, and by how much against the noise floor. One address, confirmed by a click; a stop link in every email.
Already following? Everything you follow, with a stop for each.
The company
Abnormal is the company behind Abnormal AI.ShowHide
The company
Abnormal is the company behind Abnormal AI.
ShowHideIn its own words
Stated by the vendor, not checked- Positioning
- provides a leading cloud-native email security platform that leverages AI-based behavioral data science to stop socially-engineered and never-seen-before email attacks that evade traditional secure email gateways
- For
- finance teams
- Customers named
- Valvoline
- Not stated on the pages read
- Price, starting price, free plan or trial, integrations, certifications, hosting
What Abnormal AI's own pages state, read September 28, 2026: abnormal.ai/profile/abnormal-ai, abnormal.ai/learning/black-box-ai, abnormal.ai/learning/ai-governance, abnormal.ai/learning/invoice-fraud-ai, abnormal.ai/learning/cyber-resilience-ai. A claimed page can correct any of them.
Is this your product?
Claim this page
Claiming is free and changes nothing in the data. A claimed page shows a verified contact who is told when each edition publishes and when Abnormal AI's standing changes by more than the noise floor; the right to propose corrections to the vendor table, meaning names the judge wrote that should or should not read as Abnormal AI, applied by version and listed in the change log; and a one-line description supplied by the vendor and marked as such.
What a new claim receivesHide what a new claim receives
A new claim receives the current edition's vendor brief for Abnormal AI by email, built from the raw record of the edition. It shows:
- where Abnormal AI is named, by buyer and by framing, and which cells hold its first choices;
- the claims the models make when they name it, ranked, with the strongest and the weakest quoted;
- its vocabulary against the segment leader's, and the pages the models cited;
- who was chosen in the answers that did not name Abnormal AI, and every reason the record gives;
- a battlecard for each top rival: the head-to-head split, why they win, and the reservation quoted against them;
- one page of published figures cleared to show a buyer.