| Category | Function | Share | Rank | Negative rate | Labels | Quadrant | Since September 2026 |
|---|---|---|---|---|---|---|---|
| Email security | Security operations | 32% | 1 of 76 | 6% | 51 | endorsed leader | ▲+1Since September 2026: 31% → 32%, +1 point. Inside the 11-point floor: within noise. Read over the models both editions asked. |
| AI security platforms | Security operations | 0% | 113 of 233 | 0% | 1 | under 10 labels · led by SentinelOne Singularity at 11% | |
| AI SOC agents | Security operations | 0% | 68 of 104 | 0% | 1 | under 10 labels · led by Dropzone AI at 24% |
| Category | September 2026 | Now | Change | Reading | Rank |
|---|---|---|---|---|---|
| Email security | 31% | 32% | ▲+1Since September 2026: 31% → 32%, +1 point. Inside the 11-point floor: within noise. Read over the models both editions asked. | Within noise | Rank 1 of 76, unchanged |
| AI security | 0% | 0% | newNew since September 2026: not ranked then, 0% now. | New this edition | Rank 113 of 233, unchanged |
| AI SOC agents | 0% | 0% | newNew since September 2026: not ranked then, 0% now. | New this edition | Rank 68 of 104, 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
| Model | First choice | Alternative | Mention | Negative | Labels |
|---|---|---|---|---|---|
| Claude Haiku 4.5 | 0 | 1 | 1 | 0 | 2 |
| GPT-5.4 mini | 3 | 1 | 0 | 0 | 4 |
| Gemini 3.5 Flash | 0 | 1 | 0 | 1 | 2 |
| Perplexity Sonar | 1 | 2 | 0 | 0 | 3 |
| Grok 4.20 | 1 | 1 | 1 | 1 | 4 |
| Mistral Small | 0 | 3 | 1 | 0 | 4 |
| DeepSeek V4 Flash | 2 | 3 | 0 | 0 | 5 |
| Llama 4 Maverick | 0 | 1 | 2 | 0 | 3 |
| Qwen 3.7 Flash | 1 | 4 | 0 | 0 | 5 |
| Kimi K2 | 2 | 1 | 1 | 0 | 4 |
| GLM 4.7 FlashX | 4 | 1 | 0 | 0 | 5 |
| MiniMax M2.5 | 1 | 1 | 0 | 0 | 2 |
| GPT-6 Luna | 3 | 1 | 0 | 1 | 5 |
| Muse Glimmer 30B | 0 | 5 | 0 | 0 | 5 |
Verbatim evidence the judge attached to positive labels.
“Microsoft Defender for Office 365 is the default choice for most Microsoft-365 shops—it's deeply integrated, cost-effective, and provides a solid baseline.” GLM 4.7 FlashX · Email security · comparative prompt · first choice
“Start with what you already have — if you're on Microsoft 365 E5, check if Defender for O365 Plan 2 is already included before paying for a new platform.” DeepSeek V4 Flash · Email security · scale prompt · first choice
“Microsoft Defender for Office 365 is the most pragmatic starting point—low cost, deep integration, and solid baseline protection.” GLM 4.7 FlashX · Email security · direct prompt · first choice
“Bottom line: For most budget-conscious businesses, Microsoft Defender for Office 365 Plan 1 at $2/user/month is the best value” DeepSeek V4 Flash · Email security · budget prompt · first choice
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.
“Microsoft's published comparison, for example, uses different "missed threat" criteria for Defender than for competing gateways, so it shouldn't be treated as a neutral head-to-head ranking.” GPT-6 Luna · Email security · negative prompt · soft negative
“strong execution but limited vision for evolving threats. Complaints include high false positives, poor internal email scanning” Grok 4.20 · Email security · negative prompt · soft negative
“relying *solely* on native cloud email security is a risk for organizations with high threat profiles” Gemini 3.5 Flash · Email security · negative prompt · soft negative
Citations exist only for the models that return a source list, five of the fourteen in this edition, so these counts come from 163 of the 176 answers that named Microsoft Defender for Office 365 and are not a share of its labels.
579 of the 579 domain citations in answers naming Microsoft Defender for Office 365 came from somebody else's page.
Pages are listed as the models cited them.
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.
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.
What Microsoft Defender for Office 365's own pages state, read September 28, 2026: microsoft.com/bg-bg/microsoft-365/exchange/office-365-message-encryption, microsoft.com/bg-bg/microsoft-365/exchange/exchange-email-security-spam-protection, microsoft.com/de-de/microsoft-365/exchange/exchange-email-security-spam-protection, microsoft.com/es-es/microsoft-365/exchange/exchange-email-security-spam-protection, microsoft.com/fr-fr/microsoft-365/exchange/exchange-email-security-spam-protection. A claimed page can correct any of them.
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 Microsoft Defender for Office 365'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 Microsoft Defender for Office 365, applied by version and listed in the change log; and a one-line description supplied by the vendor and marked as such.
A new claim receives the current edition's vendor brief for Microsoft Defender for Office 365 by email, built from the raw record of the edition. It shows:
AI models recommend Microsoft Defender for Office 365 first in Email security this edition. Each badge says so in the buyer's words, names the edition, and links to the standing. The next edition issues a new badge; this one stays true as a record of October 2026.
Badges read on light and dark pages. Alt text carries the claim, the category, the edition and the index, so it stays a citation where the image does not load.