Seven of fourteen models named Microsoft Purview DLP first on the direct prompt; zero named Forcepoint DLP. Microsoft Purview DLP was named by thirteen of the fourteen models and Forcepoint DLP by thirteen and Microsoft Purview DLP carries 56 labels and Forcepoint DLP 43, so the shares are not directly comparable.
Named in five categories this edition.
Named in three categories this edition.
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; rank is within the category; every quote names the model and the prompt it came from. Both figures come from the data loss prevention page.
Across every category in the October 2026 Edition, Microsoft Purview DLP and Forcepoint DLP were named in the same answer 120 times, of the 187 answers naming Microsoft Purview DLP and the 170 naming Forcepoint DLP. In those answers Forcepoint DLP took the first choice nine times and Microsoft Purview DLP fifty-nine.
| Model | Direct | Paraphrase | Comparative | Budget-constrained | Scale-constrained | Negative |
|---|---|---|---|---|---|---|
| Claude Haiku 4.5 | ||||||
| GPT-5.4 mini | ||||||
| Gemini 3.5 Flash | ||||||
| Perplexity Sonar | ||||||
| Grok 4.1 Fast | ||||||
| Mistral Small | ||||||
| DeepSeek V4 Flash | ||||||
| Llama 4 Maverick | ||||||
| Qwen 3.7 Flash | ||||||
| Kimi K2 | ||||||
| GLM 4.7 FlashX | ||||||
| MiniMax M2.5 | ||||||
| GPT-6 Luna | ||||||
| Muse Glimmer 30B |
Bold names in an answer are the products the judge labeled a first choice; a model naming several gives each of them that label. The full answer text for every row is in the record.
Every negative label with a quote, up to three, then the highest-weighted positives, up to three. Six of seven in this category shown.
“When to avoid: Non-Microsoft ecosystems or diverse file types/GenAI risks.” Grok 4.1 Fast · negative prompt · hard negative
“Microsoft Purview is highly effective if your organization is completely standardized on the Windows and Microsoft 365 ecosystem. However, you should be cautious if your environment is heterogeneous.” Gemini 3.5 Flash · negative prompt · soft negative
“Common caution themes in recent reviews include missing basic capabilities for some use cases, inconsistent design across workloads, and support that reviewers describe as slow or inflexible.” GPT-5.4 mini · negative prompt · soft negative
“Best value if you already pay for Microsoft 365... Microsoft Purview DLP is a no-brainer—no new infrastructure, just policy configuration.” GLM 4.7 FlashX · budget prompt · first choice
“If you use Microsoft 365: Start with Microsoft Purview DLP for Exchange, SharePoint, and OneDrive if your licenses include it.” GPT-6 Luna · budget prompt · first choice
“The top 5 Data Loss Prevention solutions are Microsoft Purview Data Loss Prevention... best for teams standardized on Microsoft 365” Claude Haiku 4.5 · comparative prompt · first choice
Every negative label with a quote, up to three, then the highest-weighted positives, up to three. Six of eight in this category shown.
“rarely the heavy, legacy enterprise suite (like Broadcom/Symantec or traditional Forcepoint). Those legacy tools are notorious for requiring multi-month deployments” Gemini 3.5 Flash · direct prompt · hard negative
“When to avoid: Large-scale hybrid environments or if you need low-maintenance AI-driven detection.” Grok 4.1 Fast · negative prompt · hard negative
“What to Avoid for Tight Budgets - Enterprise suites (Forcepoint, Symantec, Proofpoint)” DeepSeek V4 Flash · budget prompt · hard negative
“or Forcepoint DLP if you need a unified policy engine that works across cloud, email, endpoints and on-premises systems” GLM 4.7 FlashX · paraphrase prompt · first choice
“Proofpoint and Forcepoint are top-tier choices due to their comprehensive features and scalability” Mistral Small · paraphrase prompt · first choice
“Forcepoint if you need stronger classic DLP controls across endpoints and web traffic” Perplexity Sonar · paraphrase prompt · first choice
Comparisons are drawn for the top eight products in each category, each against each. The output is the models' output; nothing here is a recommendation by the index.