AI security platforms
Asked as “AI security platform”, and as “tool to secure LLM applications and employee AI use”, on behalf of a mid-market B2B company. 51 first choices recorded across the direct, paraphrase, budget and scale prompts, fourteen models each. Added to the November 2026 Edition.
Standing · first-choice share
12%
Contested · SentinelOne Singularity 8%
12Microsoft Defender for Cloud08SentinelOne Singularity06Microsoft 36575others
12% of first choices, contested.
Since October 2026↗new leaderNew leader since October 2026: Microsoft Defender for Cloud (12%) replaces SentinelOne Singularity (11% then, 8% now), 4 points clear, inside the 11-point floor.Microsoft Defender for Cloud leads at 12%, replacing SentinelOne Singularity, which led at 11% and stands at 8% now: 4 points clear, inside the floor, so the swap reads as unsettled.
By buyer segment
The same question asked on behalf of a different buyer. Each standing is computed within its segment; they sit side by side and are never added together.
Small business
Microsoft Defender for Cloud32%Contested▼−1Since October 2026: 33% → 32%, −1 point. Inside the 11-point floor: within noise.Standing →Mid-marketMicrosoft Defender for Cloud12%Contested↗new leaderNew leader since October 2026: Microsoft Defender for Cloud (12%) replaces SentinelOne Singularity (11% then, 8% now), 4 points clear, inside the 11-point floor.This pageEnterprise
Check Point AI Security12%Contested▲+1Since October 2026: 11% → 12%, +1 point. Inside the 11-point floor: within noise.Standing →
Microsoft Defender for Cloud32%Contested▼−1Since October 2026: 33% → 32%, −1 point. Inside the 11-point floor: within noise.Standing →Mid-marketMicrosoft Defender for Cloud12%Contested↗new leaderNew leader since October 2026: Microsoft Defender for Cloud (12%) replaces SentinelOne Singularity (11% then, 8% now), 4 points clear, inside the 11-point floor.This pageEnterpriseCheck Point AI Security12%Contested▲+1Since October 2026: 11% → 12%, +1 point. Inside the 11-point floor: within noise.Standing →The standing
Share is the count of first choices across the direct, paraphrase, budget and scale prompts, over all fourteen models, for a mid-market B2B company. Ordered by share.
ProductFirst-choice shareNegative rateLabelsQuadrantSince October 2026
01Microsoft Defender for Cloud12%0%10accepted challenger▲+1Since October 2026: 11% → 12%, +1 point. Inside the 11-point floor: within noise.11% → 12%
Microsoft Defender for Cloud12%0%10accepted challenger▲+1Since October 2026: 11% → 12%, +1 point. Inside the 11-point floor: within noise.11% → 12%02
SentinelOne Singularity8%0%17accepted challenger▼−3Since October 2026: 11% → 8%, −3 points. Inside the 11-point floor: within noise.11% → 8%
SentinelOne Singularity8%0%17accepted challenger▼−3Since October 2026: 11% → 8%, −3 points. Inside the 11-point floor: within noise.11% → 8%03
Microsoft Purview6%0%11accepted challenger▲+2Since October 2026: 4% → 6%, +2 points. Inside the 11-point floor: within noise.4% → 6%
Microsoft Purview6%0%11accepted challenger▲+2Since October 2026: 4% → 6%, +2 points. Inside the 11-point floor: within noise.4% → 6%04
Lakera Guard4%0%17accepted challenger▲+4Since October 2026: 0% → 4%, +4 points. Inside the 11-point floor: within noise.0% → 4%
Lakera Guard4%0%17accepted challenger▲+4Since October 2026: 0% → 4%, +4 points. Inside the 11-point floor: within noise.0% → 4%05Check Point AI Security4%12%17accepted challenger=heldSince October 2026: 4% → 4%, ±0 points. Inside the 11-point floor: within noise.4% → 4%
Check Point AI Security4%12%17accepted challenger=heldSince October 2026: 4% → 4%, ±0 points. Inside the 11-point floor: within noise.4% → 4%06
CrowdStrike Falcon2%0%14accepted challenger▼−2Since October 2026: 4% → 2%, −2 points. Inside the 11-point floor: within noise.4% → 2%
CrowdStrike Falcon2%0%14accepted challenger▼−2Since October 2026: 4% → 2%, −2 points. Inside the 11-point floor: within noise.4% → 2%Show the three products at 0%, ordered by negative rate
09
Darktrace0%27%15criticized challenger=heldSince October 2026: 0% → 0%, ±0 points. Inside the 11-point floor: within noise.0% → 0%
Darktrace0%27%15criticized challenger=heldSince October 2026: 0% → 0%, ±0 points. Inside the 11-point floor: within noise.0% → 0%08Microsoft Security Copilot0%18%11accepted challenger=heldSince October 2026: 0% → 0%, ±0 points. Inside the 11-point floor: within noise.0% → 0%
07
HiddenLayer0%0%14accepted challenger=heldSince October 2026: 0% → 0%, ±0 points. Inside the 11-point floor: within noise.0% → 0%
HiddenLayer0%0%14accepted challenger=heldSince October 2026: 0% → 0%, ±0 points. Inside the 11-point floor: within noise.0% → 0%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 can't be told apart from repeat variation, and both are shown. Each index measures its own floor from its own repeat, over its own categories, so the floor differs from one index to the next. Both editions asked the same fourteen models. How the floor is measured
Bars are the share of first choices, 0 to 100Every product with at least 10 labels here. Every product name links to its product page.
Head to head
All fifteen head-to-head pages: the top six products, each against each
Microsoft Defender for Cloud vs Check Point AI SecurityMicrosoft Defender for Cloud vs CrowdStrike FalconSentinelOne Singularity vs Microsoft PurviewSentinelOne Singularity vs Lakera GuardSentinelOne Singularity vs Check Point AI SecuritySentinelOne Singularity vs CrowdStrike FalconMicrosoft Purview vs Lakera GuardMicrosoft Purview vs Check Point AI SecurityMicrosoft Purview vs CrowdStrike FalconLakera Guard vs Check Point AI SecurityLakera Guard vs CrowdStrike FalconCheck Point AI Security vs CrowdStrike Falcon
Recommended versus criticized
Every product with at least 10 labels here, on both axes. The 30% line names a quadrant, not the verdict above: that one needs more than 40%.
Criticized challengerCriticized default
Negative label rate →
01020304050607M08
09Accepted challengerEndorsed leader
0%First-choice share → · lines at 30% share and 25% negative30%
Key
01Microsoft Defender for Cloud12%
Microsoft Defender for Cloud12%02SentinelOne Singularity8%
SentinelOne Singularity8%03Microsoft Purview6%
Microsoft Purview6%04Lakera Guard4%
Lakera Guard4%05Check Point AI Security4%
Check Point AI Security4%06CrowdStrike Falcon2%
CrowdStrike Falcon2%07HiddenLayer0%
HiddenLayer0%08Microsoft Security Copilot0%
09Darktrace0%
Darktrace0%What they warned about
Zero of fourteen models held their first choice under the paraphrase. Claude Haiku 4.5, GPT-5.4 mini, Gemini 3.5 Flash, Perplexity Sonar, Grok 4.20, Mistral Small, DeepSeek V4 Flash, Llama 4 Maverick, Qwen 3.7 Flash, Kimi K2, GLM 4.7 FlashX, MiniMax M2.5, GPT-6 Luna and Muse Glimmer 30B changed. A high negative share on a product with few labels is a warning. A low share on a product with many labels is salience, not sentiment.
Darktrace27%
4 of 15 labels negative · 4 of 14 models · 1 hard negative
“some models advise avoiding buying a sprawling enterprise platform like Darktrace” Llama 4 Maverick, negative prompt
DeepSeek100%
3 of 3 labels negative · 3 of 14 models · 2 hard negative
“You should be particularly cautious about DeepSeek... sending user prompts and data to Chinese entities... risky for enterprise use.” GLM 4.7 FlashX, negative prompt
ChatGPT100%
3 of 3 labels negative · 3 of 14 models · 1 hard negative
“The free version of ChatGPT does not guarantee data privacy... This makes it unsuitable for enterprise use” Mistral Small, negative prompt
Google Gemini
100%
3 of 3 labels negative · 3 of 14 models · 1 hard negative
“Google Gemini or related products (high risk/D): Leverages Google's vast data ecosystem, raising privacy and lock-in concerns.” Grok 4.20, negative prompt
What they cite
Fourteen of the fourteen models returned a source list in this edition: five return the pages they cited and nine the search results they consulted.
Sites the answers cite
74 of 84 answers in this category came back with a source list, from 14 of 14 models: citations where the model returns them, or the search results it consulted. 802 links across 299 sites, every framing counted. Ranked by the number of answers carrying the site or page. 46 of the 252 answers across every segment cited this index's own page for the category; the method page measures whether that reading tilts an answer.
vendor site · Check Point34 answers · 39 citations · 12 models
23 answers · 34 citations · 9 models
vendor site · Microsoft17 answers · 30 citations · 8 models
12 answers · 12 citations · 6 models
vendor site · Superblocks12 answers · 12 citations · 8 models
publisher (Gane)11 answers · 19 citations · 8 models
vendor site · Rework11 answers · 11 citations · 10 models
vendor site · Cyera10 answers · 15 citations · 6 models
vendor site · Akamai10 answers · 13 citations · 5 models
vendor site · Venn10 answers · 12 citations · 4 models
vendor site · Witness.ai9 answers · 14 citations · 7 models
vendor site · Forcepoint9 answers · 10 citations · 7 models
Pages the answers cite
The ten pages named in the most answers, by full address. A page here is one the models returned with a recommendation, not one the index endorses.
23 answers · 12 models
11 answers · 10 models
9 answers · 6 models
9 answers · 3 models
8 answers · 7 models
7 answers · 5 models
7 answers · 6 models
7 answers · 4 models
7 answers · 6 models
7 answers · 5 models
Search against answers
Each company's standing in the answers beside its site's footprint in Google search, one row a site: the products the models named on it with their shares, and the share they add up to; monthly searches on Google, and DataForSEO's estimate of AI search demand (modeled from search signals, directional, not a count of queries to any assistant), for the most-searched of the company's and its products' names (the name is in each row's hover text); estimated monthly organic visits to the site; and its best position in Google's top ten for “ai security platform”, “ai security platforms”, “best ai security platform”. US estimates from DataForSEO and Google's Ads Transparency Center. A small company's site, or a mid-sized company's site for its flagship, is marked company; a product on a large parent's site (Google, Microsoft) has no site figures. A column with no figures for this category is left out, and an empty cell means none were seen, not none exist. Two measurements side by side: neither is read as the cause of the other.
| Company and products | Share | Own site cited | Name searches, Google | AI search demand, est. | Organic visits | Paid search a month, est. | Google ads, last 30 days | Best Google position |
|---|---|---|---|---|---|---|---|---|
| Microsoft Defender microsoft.com Microsoft Defender for Cloud 12% | 12% | 659 | 18,100 | 3,525 | 145,358,109 company | $12,455 | 103 | |
| Microsoft parent site Microsoft Purview 6%, Microsoft 365 6% | 12% | 514 | 2,240,000 | 403,136 | ||||
| SentinelOne sentinelone.com SentinelOne Singularity 8% | 8% | 499 | 33,100 | 1,675 | 446,702 company | $780 | 22 | |
| CrowdStrike parent site CrowdStrike Falcon Go 4%, CrowdStrike Falcon 2% | 6% | 201 | 74,000 | 5,130 | ||||
| Lakera Guard lakera.ai | 4% | 44 | 170 | 3 | 2,639 | |||
| Check Point parent site Check Point AI Security 4% | 4% | 46 | 14,800 | 4,487 | ||||
| Microsoft Defender parent site Microsoft Defender XDR 4% | 4% | 199 | 18,100 | 3,525 | ||||
| Prompt Security prompt.security | 4% | 30 | 1,000 | 41 | 2,008 | |||
| dope.security dope.security | 4% | 24 | ||||||
| WitnessAI witness.ai | 4% | 25 | 590 | 14 | 3,200 | 1 | 3 |
Google's top ten for the three searches hold 18 sites; 2 of them are among the sites the models cited here (cyera.com, witness.ai). The first result for “ai security platform” is reddit.com.
How they answered
Six framings of the same buying question, each sent to every model in a fresh session with search on. One row per model, so a row shows whether it held its answer under rewording, what it named when cost was the constraint, and what it argued against. Computed from the raw judge labels.ShowHide
How they answered
Six framings of the same buying question, each sent to every model in a fresh session with search on. One row per model, so a row shows whether it held its answer under rewording, what it named when cost was the constraint, and what it argued against. Computed from the raw judge labels.
ShowHideModelDirect“What is the best AI security platform for a mid-market B2B company?”Paraphrase“Which tool to secure LLM applications and employee AI use would you recommend to a mid-sized B2B company?”Comparative“What are the top AI security platforms and how do they differ?”Budget-constrained“What is the best AI security platform for a company with a limited budget?”Scale-constrained“We are a 500 person company evaluating an AI security platform. What should we look at?”Negative“Which AI security platforms should I avoid or be cautious about?”
Claude Haiku 4.5no first choicePrompt Security, WitnessAIChangedno first choiceno first choiceno first choicenothing named
Two alternatives
Lasso Security, Nudge SecurityGPT-5.4 miniOrca Securityno first choiceChangedPalo Alto Networks Cortex Cloud, Palo Alto Networks Prisma AIRSWazuhagainst: CrowdStrike Falcon Cloud Security, Varonis, Wizno first choicenothing named
Four alternatives
CrowdStrike, Palo Alto Networks, Securiti, WizFive alternatives
Check Point AI Security, CrowdStrike Falcon, Darktrace, Orca Security, Zscaler AI security / AI GuardFour alternatives
Microsoft Defender for Cloud, Orca Security, Securiti, UTMStackGemini 3.5 FlashLakera Guard, SentinelOne SingularityPrompt SecurityChangedno first choiceGarak, NVIDIA NeMo Guardrailsno first choicenothing named
Seven alternatives
Aikido Security, Charlotte AI, CrowdStrike Falcon, Dropzone AI, Harmonic Security, Nightfall AI, SeceonFive alternatives
Cloudflare One, Defender for Cloud Apps, Lakera Guard, Microsoft Purview, Netskope One AI SecuritySix alternatives
Check Point – Lakera, Cisco AI Defense, F5 – CalypsoAI, HiddenLayer, Palo Alto Networks – Prisma AIRS, SentinelOne – Prompt SecuritySix alternatives
Amazon Bedrock Guardrails, Azure AI Content Safety, Guardrails AI, Langfuse, Meta Llama Guard, PromptfooPerplexity SonarMicrosoft Defender XDRagainst: Palo Alto Networks Prisma AIRSPrompt Security—now part of SentinelOneChangedPalo Alto Networks Prisma AIRSHuntress Managed EDRagainst: Wazuhno first choiceagainst: DeepSeek
Six alternatives
Cisco AI Defense, CrowdStrike Falcon with AIDR/Charlotte AI, Cyera, Lakera Guard, Microsoft Purview, Sophos Intercept XFour alternatives
Cisco AI Defense, Lakera Guard, Microsoft Purview, Prisma AIRSThirteen alternatives
Checkmarx One, Cisco AI Defense, CrowdStrike Falcon, Cyera, Defender, HiddenLayer, Lakera Guard, Microsoft Purview, Microsoft Security, SentinelOne Singularity, Varonis, Wiz, ZenityThree alternatives
Cynet, Lakera Guard, Microsoft Defender XDRGrok 4.20CrowdStrike Falconagainst: Check Point AI Security, Darktrace, Vectra AICheck Point AI SecurityChangedno first choiceCrowdStrike Falcon Goagainst: Arctic Wolfno first choiceagainst: Amazon-related AI, Google Gemini, Microsoft 365 Copilot, OpenAI, Tesla Autopilot/FSD
Four alternatives
Microsoft Security Copilot, Palo Alto Networks, SentinelOne Singularity, WizSix alternatives
Azure AI Content Safety, Credo AI, Netskope One AI Security, Palo Alto Networks Prisma Access, WitnessAI, dope.securityEleven alternatives
Arctic Wolf, Cyera AI Guardian, Darktrace, Lakera Guard, Microsoft Security Copilot, Prisma AIRS, Prompt Security, Protect AI, Robust Intelligence, SentinelOne Singularity, Vectra AITwo alternatives
Microsoft 365, SentinelOne SingularityMistral SmallCrowdStrike Falcon AI Detection and Response, Nightfall AIagainst: Check Point AI Security, Microsoft Security CopilotGiskard, Lakera Guard, TigerGate AI SecurityChangedno first choiceCrowdStrike Falcon Go, Microsoft 365no first choiceagainst: ChatGPT, DeepSeek
One alternative
Noma SecurityTwo alternatives
Protect AI, Robust IntelligenceOne alternative
Cynet All-in-One Cybersecurity PlatformDeepSeek V4 FlashSentinelOne Singularitydope.securityChangedagainst: WitnessAIPalo Alto Networks Prisma AIRSMicrosoft Defender for Cloudagainst: Microsoft Sentinelno first choiceagainst: Anthropic's Claude Desktop, Cursor, GitHub Copilot, Microsoft's VS Code AI extensions, OpenAI's Codex
Four alternatives
Check Point AI Security, CrowdStrike Falcon, Microsoft Defender XDR, Security CopilotFour alternatives
Bifrost, Lakera Guard, Netskope One AI Security, Zscaler AISix alternatives
Aim Security, Cato SASE Cloud, Cisco AI Defense, HiddenLayer, Lakera Guard, WitnessAIFive alternatives
Bitdefender GravityZone, Cloudflare One, CrowdStrike Falcon Go, Guardz, Huntress Managed EDRLlama 4 MaverickSentinelOne SingularityGarakChangedno first choiceno first choiceagainst: Microsoft Security Copilotno first choiceagainst: Darktrace
Three alternatives
Bitdefender GravityZone, Microsoft Defender for Cloud, Sophos Intercept XTwo alternatives
Credo AI, Fairly AIQwen 3.7 FlashCheck Point AI SecurityGoogle DLP, Microsoft PurviewChangedForcepoint AI Security, LayerX SecurityMicrosoft Defender for Cloudno first choiceagainst: AIDR, CrowdStrike, Darktrace, PentestGPT, Prophet Security, Vectra AI
Five alternatives
CrowdStrike, Fortinet, Microsoft Security Copilot, UnderDefense AI SOC, dope.securityFour alternatives
Cisco Secure Endpoint, Prevalent, Vermilion, ZscalerSix alternatives
Aim Security, Check Point AI Security, Cisco, Prompt Security, Robust Intelligence, Talon SecurityTwo alternatives
CrowdStrike Falcon Go, CynetKimi K2SentinelOne SingularityHarmonic SecurityChangedCrowdStrike Falcon, Darktrace, SentinelOne SingularityMalwarebytes for Businessno first choiceagainst: CrowdStrike, Darktrace, Evolv Technologies, LiteLLM
Four alternatives
CrowdStrike Falcon, Microsoft Defender XDR, Palo Alto Networks Cortex XDR, Security CopilotFour alternatives
Check Point AI Security, Cloudflare AI Security Suite, Microsoft Purview, SentinelOne Prompt SecuritySix alternatives
Cyera, HiddenLayer, Microsoft Security Copilot, Palo Alto Networks Cortex XDR, Robust Intelligence, Vectra AIThree alternatives
CrowdStrike Falcon Go, Microsoft Defender for Cloud, SentinelOne SingularityGLM 4.7 FlashXMicrosoft Defender XDRWalled AIChangedMicrosoft Security CopilotMicrosoft Defender for Cloudno first choiceagainst: Anthropic's Claude Code, Anthropic's Claude Desktop, ChatGPT, Cursor, DeepSeek, GitHub Copilot, Google Gemini, OpenAI's Codex, The Ray Framework
Six alternatives
CrowdStrike Falcon, Fortinet, HiddenLayer, Microsoft Purview, Mindgard, dope.securityFive alternatives
Akamai Workforce Protector, LangProtect, Microsoft Purview, QuilrAI, RepacketFour alternatives
CrowdStrike Falcon, Darktrace, Forcepoint AI Data Security, SentinelOne SingularityThree alternatives
CrowdStrike Falcon Go, Lakera Guard, NVIDIA NeMo GuardrailsMiniMax M2.5CrowdStrike Falcon AI Detection and Response, Microsoft Defender for CloudMicrosoft PurviewChangedno first choiceMicrosoft 365, Microsoft Defender for Cloudagainst: Arctic Wolfno first choiceagainst: COMPAS, Clearview AI, Rekognition
Two alternatives
Fortinet Security Fabric, Noma SecurityFive alternatives
Credo AI, Fairly AI, Netskope Skylight Ai, Venn Blue Border, WitnessAITwo alternatives
CrowdStrike Falcon Go, Microsoft Defender XDRGPT-6 LunaMicrosoft PurviewCloudflare OneChangedCisco AI Defense, Palo Alto Networks Prisma AIRSGoogle Cloud Model Armorno first choicenothing named
Two alternatives
Harmonic Security, Proofpoint AI SecurityTwo alternatives
Microsoft Purview DSPM for AI and Defender for Cloud Apps, Netskope One AI SecurityFour alternatives
Amazon Bedrock Guardrails, Check Point AI Security, including Lakera, Google Model Armor, HiddenLayerOne alternative
Cloudflare AI GatewayMuse Glimmer 30BUnderDefense AI SOCWitnessAI, dope.securityChangedCisco, Microsoft Security CopilotHuntress Managed EDR, Microsoft 365, Microsoft Defender for Cloudno first choiceagainst: Anthropic's Claude Desktop, ChatGPT, Claude, Cursor, GitHub, GitHub Copilot, Google Gemini, Hugging Face, Microsoft 365 Copilot, Microsoft's VS Code, OpenAI
Nine alternatives
Darktrace, Daylight Security, Dropzone AI, Echelon Risk + Cyber, Guardz, Microsoft Security Copilot, Prophet Security, Seceon, Vectra AIOne alternative
Akamai's Workforce ProtectorSeven alternatives
Check Point AI Security, CrowdStrike Falcon, HiddenLayer, Lakera Guard, Protect AI, Robust Intelligence, Wiz AI-APPTwo alternatives
Microsoft Security Copilot, Nightfall AITwo alternatives
Azure AI Content Safety, Protectt.aiBold is the first choiceAlternatives are counted; the count opens them.What the answer argued against
The record
One row per call: the version string exactly as returned, whether the model searched, sources cited, and latency.
Eighty-four rows: every prompt, every model, every answer.
| Prompt | Model | Version string | Time (UTC) | Searched | Sources | Latency |
|---|---|---|---|---|---|---|
| Direct recommendation | Claude Haiku 4.5 | claude-haiku-4-5-20251001 | 2026-11-01 12:10 | no | 0 | 4 s |
| Direct recommendation | GPT-5.4 mini | gpt-5.4-mini-2026-03-17 | 2026-11-01 08:47 | yes | 2 | 5 s |
| Direct recommendation | Gemini 3.5 Flash | gemini-3.6-flash (asked for gemini-3.5-flash; the API answered as this version) | 2026-11-01 09:12 | yes | 13 | 23 s |
| Direct recommendation | Perplexity Sonar | sonar | 2026-11-01 12:38 | yes | 10 | 4 s |
| Direct recommendation | Grok 4.20 | spacexai/grok-4.20-non-reasoning via xai | 2026-11-01 19:22 | yes | 9 | 12 s |
| Direct recommendation | Mistral Small | mistral/mistral-small via mistral | 2026-11-01 13:49 | yes | 5 | 4 s |
| Direct recommendation | DeepSeek V4 Flash | deepseek/deepseek-v4-flash via deepinfra | 2026-11-01 17:56 | yes | 22 | 23 s |
| Direct recommendation | Llama 4 Maverick | meta/llama-4-maverick via bedrock | 2026-11-01 18:45 | yes | 5 | 2 s |
| Direct recommendation | Qwen 3.7 Flash | alibaba/qwen3.7-flash via alibaba | 2026-11-01 12:37 | yes | 5 | 13 s |
| Direct recommendation | Kimi K2 | moonshotai/kimi-k2 via novita | 2026-11-01 17:15 | yes | 20 | 33 s |
| Direct recommendation | GLM 4.7 FlashX | zai/glm-4.7-flashx via zai | 2026-11-01 11:06 | yes | 9 | 26 s |
| Direct recommendation | MiniMax M2.5 | minimax/minimax-m2.5 via minimax | 2026-11-01 10:59 | yes | 9 | 34 s |
| Direct recommendation | GPT-6 Luna | gpt-6-luna | 2026-11-01 09:39 | yes | 3 | 14 s |
| Direct recommendation | Muse Glimmer 30B | meta/muse-glimmer-30b via togetherai | 2026-11-01 12:28 | yes | 12 | 21 s |
| Paraphrase | Claude Haiku 4.5 | claude-haiku-4-5-20251001 | 2026-11-01 12:44 | yes | 17 | 10 s |
| Paraphrase | GPT-5.4 mini | gpt-5.4-mini-2026-03-17 | 2026-11-01 10:56 | yes | 3 | 6 s |
| Paraphrase | Gemini 3.5 Flash | gemini-3.6-flash (asked for gemini-3.5-flash; the API answered as this version) | 2026-11-01 17:24 | yes | 8 | 21 s |
| Paraphrase | Perplexity Sonar | sonar | 2026-11-01 15:41 | yes | 10 | 4 s |
| Paraphrase | Grok 4.20 | spacexai/grok-4.20-non-reasoning via xai | 2026-11-01 18:00 | yes | 15 | 15 s |
| Paraphrase | Mistral Small | mistral/mistral-small via mistral | 2026-11-01 09:06 | yes | 5 | 5 s |
| Paraphrase | DeepSeek V4 Flash | deepseek/deepseek-v4-flash via deepinfra | 2026-11-01 09:34 | yes | 24 | 21 s |
| Paraphrase | Llama 4 Maverick | meta/llama-4-maverick via bedrock | 2026-11-01 09:03 | yes | 5 | 3 s |
| Paraphrase | Qwen 3.7 Flash | alibaba/qwen3.7-flash via alibaba | 2026-11-01 10:55 | no | 0 | 31 s |
| Paraphrase | Kimi K2 | moonshotai/kimi-k2 via novita | 2026-11-01 10:37 | yes | 20 | 45 s |
| Paraphrase | GLM 4.7 FlashX | zai/glm-4.7-flashx via zai | 2026-11-01 10:49 | yes | 14 | 55 s |
| Paraphrase | MiniMax M2.5 | minimax/minimax-m2.5 via minimax | 2026-11-01 15:04 | yes | 5 | 25 s |
| Paraphrase | GPT-6 Luna | gpt-6-luna | 2026-11-01 14:55 | yes | 3 | 22 s |
| Paraphrase | Muse Glimmer 30B | meta/muse-glimmer-30b via togetherai | 2026-11-01 17:06 | yes | 14 | 19 s |
| Comparative | Claude Haiku 4.5 | claude-haiku-4-5-20251001 | 2026-11-01 15:51 | yes | 9 | 10 s |
| Comparative | GPT-5.4 mini | gpt-5.4-mini-2026-03-17 | 2026-11-01 13:43 | yes | 6 | 9 s |
| Comparative | Gemini 3.5 Flash | gemini-3.6-flash (asked for gemini-3.5-flash; the API answered as this version) | 2026-11-01 09:55 | yes | 10 | 24 s |
| Comparative | Perplexity Sonar | sonar | 2026-11-01 15:01 | yes | 10 | 7 s |
| Comparative | Grok 4.20 | spacexai/grok-4.20-non-reasoning via xai | 2026-11-01 16:54 | yes | 5 | 16 s |
| Comparative | Mistral Small | mistral/mistral-small via mistral | 2026-11-01 11:06 | yes | 5 | 7 s |
| Comparative | DeepSeek V4 Flash | deepseek/deepseek-v4-flash via deepinfra | 2026-11-01 17:01 | yes | 24 | 33 s |
| Comparative | Llama 4 Maverick | meta/llama-4-maverick via bedrock | 2026-11-01 17:50 | yes | 5 | 3 s |
| Comparative | Qwen 3.7 Flash | alibaba/qwen3.7-flash via alibaba | 2026-11-01 17:22 | yes | 5 | 37 s |
| Comparative | Kimi K2 | moonshotai/kimi-k2 via novita | 2026-11-01 17:21 | yes | 19 | 52 s |
| Comparative | GLM 4.7 FlashX | zai/glm-4.7-flashx via zai | 2026-11-01 13:54 | yes | 24 | 189 s |
| Comparative | MiniMax M2.5 | minimax/minimax-m2.5 via minimax | 2026-11-01 16:21 | yes | 10 | 14 s |
| Comparative | GPT-6 Luna | gpt-6-luna | 2026-11-01 16:19 | yes | 7 | 24 s |
| Comparative | Muse Glimmer 30B | meta/muse-glimmer-30b via togetherai | 2026-11-01 19:35 | yes | 14 | 25 s |
| Budget constrained | Claude Haiku 4.5 | claude-haiku-4-5-20251001 | 2026-11-01 13:46 | no | 0 | 4 s |
| Budget constrained | GPT-5.4 mini | gpt-5.4-mini-2026-03-17 | 2026-11-01 09:33 | yes | 2 | 4 s |
| Budget constrained | Gemini 3.5 Flash | gemini-3.6-flash (asked for gemini-3.5-flash; the API answered as this version) | 2026-11-01 12:03 | yes | 5 | 23 s |
| Budget constrained | Perplexity Sonar | sonar | 2026-11-01 11:54 | yes | 10 | 3 s |
| Budget constrained | Grok 4.20 | spacexai/grok-4.20-non-reasoning via xai | 2026-11-01 14:49 | yes | 5 | 7 s |
| Budget constrained | Mistral Small | mistral/mistral-small via mistral | 2026-11-01 10:37 | yes | 5 | 4 s |
| Budget constrained | DeepSeek V4 Flash | deepseek/deepseek-v4-flash via deepinfra | 2026-11-01 16:55 | yes | 25 | 31 s |
| Budget constrained | Llama 4 Maverick | meta/llama-4-maverick via bedrock | 2026-11-01 14:05 | yes | 5 | 3 s |
| Budget constrained | Qwen 3.7 Flash | alibaba/qwen3.7-flash via alibaba | 2026-11-01 16:08 | yes | 5 | 29 s |
| Budget constrained | Kimi K2 | moonshotai/kimi-k2 via novita | 2026-11-01 17:01 | yes | 9 | 13 s |
| Budget constrained | GLM 4.7 FlashX | zai/glm-4.7-flashx via zai | 2026-11-01 09:05 | yes | 14 | 17 s |
| Budget constrained | MiniMax M2.5 | minimax/minimax-m2.5 via minimax | 2026-11-01 10:50 | yes | 5 | 46 s |
| Budget constrained | GPT-6 Luna | gpt-6-luna | 2026-11-01 19:29 | yes | 2 | 17 s |
| Budget constrained | Muse Glimmer 30B | meta/muse-glimmer-30b via togetherai | 2026-11-01 18:57 | yes | 14 | 24 s |
| Scale constrained | Claude Haiku 4.5 | claude-haiku-4-5-20251001 | 2026-11-01 11:05 | no | 0 | 5 s |
| Scale constrained | GPT-5.4 mini | gpt-5.4-mini-2026-03-17 | 2026-11-01 16:52 | no | 0 | 7 s |
| Scale constrained | Gemini 3.5 Flash | gemini-3.6-flash (asked for gemini-3.5-flash; the API answered as this version) | 2026-11-01 15:24 | no | 0 | 17 s |
| Scale constrained | Perplexity Sonar | sonar | 2026-11-01 15:19 | yes | 10 | 8 s |
| Scale constrained | Grok 4.20 | spacexai/grok-4.20-non-reasoning via xai | 2026-11-01 19:17 | yes | 10 | 15 s |
| Scale constrained | Mistral Small | mistral/mistral-small via mistral | 2026-11-01 16:34 | no | 0 | 7 s |
| Scale constrained | DeepSeek V4 Flash | deepseek/deepseek-v4-flash via deepinfra | 2026-11-01 15:44 | yes | 22 | 54 s |
| Scale constrained | Llama 4 Maverick | meta/llama-4-maverick via bedrock | 2026-11-01 14:22 | yes | 5 | 3 s |
| Scale constrained | Qwen 3.7 Flash | alibaba/qwen3.7-flash via alibaba | 2026-11-01 17:09 | no | 0 | 33 s |
| Scale constrained | Kimi K2 | moonshotai/kimi-k2 via novita | 2026-11-01 12:04 | yes | 10 | 55 s |
| Scale constrained | GLM 4.7 FlashX | zai/glm-4.7-flashx via zai | 2026-11-01 18:17 | yes | 15 | 21 s |
| Scale constrained | MiniMax M2.5 | minimax/minimax-m2.5 via minimax | 2026-11-01 16:52 | no | 0 | 18 s |
| Scale constrained | GPT-6 Luna | gpt-6-luna | 2026-11-01 16:00 | yes | 2 | 19 s |
| Scale constrained | Muse Glimmer 30B | meta/muse-glimmer-30b via togetherai | 2026-11-01 14:15 | yes | 15 | 29 s |
| Negative framing | Claude Haiku 4.5 | claude-haiku-4-5-20251001 | 2026-11-01 13:27 | yes | 19 | 11 s |
| Negative framing | GPT-5.4 mini | gpt-5.4-mini-2026-03-17 | 2026-11-01 16:28 | yes | 6 | 5 s |
| Negative framing | Gemini 3.5 Flash | gemini-3.6-flash (asked for gemini-3.5-flash; the API answered as this version) | 2026-11-01 09:45 | yes | 15 | 35 s |
| Negative framing | Perplexity Sonar | sonar | 2026-11-01 17:51 | yes | 10 | 5 s |
| Negative framing | Grok 4.20 | spacexai/grok-4.20-non-reasoning via xai | 2026-11-01 13:54 | yes | 13 | 13 s |
| Negative framing | Mistral Small | mistral/mistral-small via mistral | 2026-11-01 12:05 | yes | 5 | 5 s |
| Negative framing | DeepSeek V4 Flash | deepseek/deepseek-v4-flash via deepinfra | 2026-11-01 12:06 | yes | 24 | 25 s |
| Negative framing | Llama 4 Maverick | meta/llama-4-maverick via bedrock | 2026-11-01 11:38 | yes | 5 | 3 s |
| Negative framing | Qwen 3.7 Flash | alibaba/qwen3.7-flash via alibaba | 2026-11-01 16:34 | yes | 14 | 45 s |
| Negative framing | Kimi K2 | moonshotai/kimi-k2 via novita | 2026-11-01 11:31 | yes | 25 | 51 s |
| Negative framing | GLM 4.7 FlashX | zai/glm-4.7-flashx via zai | 2026-11-01 12:55 | yes | 23 | 215 s |
| Negative framing | MiniMax M2.5 | minimax/minimax-m2.5 via minimax | 2026-11-01 09:09 | no | 0 | 29 s |
| Negative framing | GPT-6 Luna | gpt-6-luna | 2026-11-01 16:21 | yes | 4 | 16 s |
| Negative framing | Muse Glimmer 30B | meta/muse-glimmer-30b via togetherai | 2026-11-01 13:44 | yes | 24 | 33 s |
Normalization in this category
Every judgment call made between the raw labels and the numbers above, listed so it is visible and reversible.
ShowHide
Category-scoped readings
AWS Bedrock Guardrails read as Amazon Bedrock Guardrails
Bitdefender read as Bitdefender GravityZone
Check Point read as Check Point AI Security
Check Point (CloudGuard AI) read as Check Point AI Security
Check Point (with Harmony/AI tools) read as Check Point AI Security
Check Point Software Technologies read as Check Point AI Security
Forcepoint read as Forcepoint Data Security Cloud
Huntress read as Huntress Managed EDR
Lakera read as Lakera Guard
Lakera (Lakera Guard) read as Lakera Guard
Lakera (now part of Check Point) read as Lakera Guard
Microsoft (Purview + Security Copilot) read as Microsoft Security
Microsoft Defender read as Microsoft Defender XDR
Microsoft Purview Suite for Business Premium read as Microsoft Purview
Netskope read as Netskope One AI Security
Orca read as Orca Security
Palo Alto read as Palo Alto Networks
Palo Alto (Prisma) read as Palo Alto Networks
Palo Alto Prisma AIRS read as Palo Alto Networks Prisma AIRS
SentinelOne read as SentinelOne Singularity
SentinelOne (with Singularity AI) read as SentinelOne Singularity
SentinelOne Singularity Core read as SentinelOne Singularity
Sophos read as Sophos Intercept X
UnderDefense (MAXI AI SOC) read as UnderDefense AI SOC
UnderDefense Agentic AI SOC read as UnderDefense AI SOC
Windows Defender (built-in) read as Microsoft Defender XDR
Unresolved, counted raw
Agentic Security
Akamai's Workforce Protector
Amazon-related AI (e.g., Alexa integrations or AWS AI security features)
Anthropic's Claude Code
Armorer Guard
COMPAS
Check Point AI Security, including Lakera
Check Point – Lakera (Lakera Guard)
Checkmarx One Assist
Cloudflare AI Security Suite
CrowdStrike Falcon with AIDR/Charlotte AI
Cyera AI Guardian
Cynet All-in-One Cybersecurity Platform
Daylight Security
Evolv Technologies
F5 – CalypsoAI
Google Model Armor
Granica
Harmonic
IBM QRadar Advisor with Watson
LLM Confidentiality
LLMmap
Linear Security
Mendable
Meta Llama Guard (Self-Hosted)
Microsoft Purview DSPM for AI and Defender for Cloud Apps
Microsoft's VS Code
Microsoft's VS Code AI extensions
Netskope Skylight Ai
NuGuard
Palo Alto Networks – Prisma AIRS (formerly Protect AI)
Pangea
PentestGPT
Private AI
PromptArmor
Proofpoint AI Security
Quilr
QuilrAI
Rekognition (Amazon)
SentinelOne – Prompt Security
Snyk AI Workflows
Snyk's Free plan
Talon Security
Tenable's Nessus Essentials
Tesla Autopilot/FSD
The Ray Framework (Anyscale)
TigerGate AI Security
Venn Blue Border
Vermilion
ZIVIS
Zscaler AI
Zscaler AI security / AI Guard
garak Scanner
ps-fuzz 2
Discontinued, still offered
No shut-down product was recommended here.