| Category | Function | Share | Rank | Negative rate | Labels | Quadrant | Since September 2026 |
|---|---|---|---|---|---|---|---|
| Cloud security posture management | Cloud and infrastructure | 6% | 5 of 80 | 17% | 18 | accepted challenger | =heldSince September 2026: 4% → 4%, ±0 points. Inside the 11-point floor: within noise. Read over the models both editions asked. |
| Cloud-native application protection | Cloud and infrastructure | 3% | 6 of 70 | 29% | 7 | under 10 labels · led by Wiz at 37% | ▲+2Since September 2026: 0% → 2%, +2 points. Inside the 11-point floor: within noise. Read over the models both editions asked. |
| Category | September 2026 | Now | Change | Reading | Rank |
|---|---|---|---|---|---|
| CNAPP | 0% | 2% | ▲+2Since September 2026: 0% → 2%, +2 points. Inside the 11-point floor: within noise. Read over the models both editions asked. | Within noise | Rank 47 → 6 of 70 |
| CSPM | 4% | 4% | =heldSince September 2026: 4% → 4%, ±0 points. Inside the 11-point floor: within noise. Read over the models both editions asked. | Within noise | Rank 6 → 5 of 80 |
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 | 0 | 0 | 0 | 0 |
| GPT-5.4 mini | 1 | 0 | 0 | 0 | 1 |
| Gemini 3.5 Flash | 0 | 2 | 0 | 1 | 3 |
| Perplexity Sonar | 0 | 0 | 0 | 0 | 0 |
| Grok 4.20 | 0 | 0 | 0 | 0 | 0 |
| Mistral Small | 0 | 0 | 1 | 0 | 1 |
| DeepSeek V4 Flash | 1 | 0 | 1 | 0 | 2 |
| Llama 4 Maverick | 0 | 0 | 0 | 0 | 0 |
| Qwen 3.7 Flash | 0 | 1 | 1 | 3 | 5 |
| Kimi K2 | 0 | 1 | 0 | 1 | 2 |
| GLM 4.7 FlashX | 0 | 0 | 1 | 0 | 1 |
| MiniMax M2.5 | 1 | 2 | 1 | 0 | 4 |
| GPT-6 Luna | 2 | 2 | 2 | 0 | 6 |
| Muse Glimmer 30B | 0 | 0 | 0 | 0 | 0 |
Verbatim evidence the judge attached to positive labels.
“AWS-heavy: Try AWS Security Hub CSPM. It has a 30-day trial and a perpetual free allowance” GPT-6 Luna · CSPM · budget prompt · first choice
“Start with native cloud tools (AWS Config, Security Hub, Microsoft Defender for Cloud)” MiniMax M2.5 · CNAPP · budget prompt · first choice
“AWS-only: Start with AWS Security Hub CSPM. It has a 30-day free trial” GPT-6 Luna · CNAPP · budget prompt · first choice
“Best overall for limited budget and AWS: AWS Security Hub CSPM” GPT-5.4 mini · CSPM · 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.
“Why avoid: If you are a true multi-cloud organization ... relying solely on the native tool for one provider will blindside you” Qwen 3.7 Flash · CSPM · negative prompt · hard negative
“The Trap: Relying exclusively on AWS Security Hub or Microsoft Defender for Cloud... Vendor Lock-in.” Qwen 3.7 Flash · CNAPP · negative prompt · hard negative
“Multi-cloud complexity: Avoid native-only tools.” Kimi K2 · CSPM · negative prompt · hard negative
“If you run a multi-cloud environment... using the native cloud provider tools is rarely a good idea” Gemini 3.5 Flash · CSPM · 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 74 of the 82 answers that named AWS Security Hub and are not a share of its labels.
197 of the 222 domain citations in answers naming AWS Security Hub 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 AWS Security Hub's own pages state, read September 28, 2026: aws.amazon.com/about-aws/whats-new/2025/06/aws-security-hub-risk-prioritization-response-scale. 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 AWS Security Hub'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 AWS Security Hub, 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 AWS Security Hub by email, built from the raw record of the edition. It shows: