Category
AI red teaming vendors and services
Red teaming for AI stress-tests models with adversarial prompts, prompt injections, jailbreaks, data exfiltration probes, and safety-policy circumvention. It is required or heavily encouraged by the EU AI Act (Article 15, robustness and cybersecurity), NIST AI RMF (Measure 2, adversarial testing), ISO/IEC 42001 (clause 8.3, AI system risk treatment), and the OWASP LLM Top 10.
Directory
10 AI red-teaming vendors
Ownership changes: Protect AI is now part of Palo Alto Networks; Lakera (Check Point AI Security) is now part of Check Point Software Technologies; Robust Intelligence is now part of Cisco; CalypsoAI is now part of F5; TrojAI by A10 Networks is now part of A10 Networks; ZwillGen AI Division (formerly Luminos.Law) is now part of ZwillGen. Check how each product is sold today before you buy.
| Vendor | HQ | Founded | Size | Pricing | Last checked |
|---|---|---|---|---|---|
| HiddenLayer | Austin, United States | 2022 | — | No prices on hiddenlayer.com; the site offers a demo request only. The AWS Marketplace listing shows one contract dimension at $5,000,000 per 12 months, does not define the unit and says private offers are available, so it is not a usable starting price (checked November 11, 2026). | Nov 11, 2026 |
| Protect AI | Seattle, United States | 2022 | — | Protect AI is no longer sold on its own; protectai.com redirects to Prisma AIRS. Prisma AIRS is licensed through a pool of Palo Alto Networks Software NGFW credits, with API-based runtime protection priced per billion monthly tokens. AWS Marketplace lists a Prisma AIRS API contract at $4,353.00 per 12 months without saying how many calls it covers. | Nov 19, 2026 |
| Lakera (Check Point AI Security) | Zurich, Switzerland | 2021 | — | No price list on lakera.ai: its pricing link redirects to the product console, which shows no plans. Check Point's AWS Marketplace listing (Check Point AI Defense Plane, annual subscriptions) shows: AI Agent Security Guardrails, 10 million requests a year, $74,000; AI Red Teaming, 1,000 scans a year, $52,000; Workforce AI Security, $100 per user (Enterprise) or $44 per user (Browser Only). Custom pricing is available through a private offer. Self-hosted AI Guardrails needs an Enterprise license. | Nov 11, 2026 |
| Robust Intelligence | San Francisco, United States | 2019 | — | Robust Intelligence is no longer sold on its own and its site lists no prices. Cisco sells the technology as Cisco AI Defense, also without public prices: its offer description lists Advantage, Validation Essentials and Runtime Essentials packages, bought per AI application or as committed usage. AWS Marketplace lists Cisco AI Defense by private offer, and Cisco offers a self-service red-teaming tool, AI Defense: Explorer Edition, at no upfront cost. | Nov 19, 2026 |
| CalypsoAI | Dublin, Ireland | 2018 | — | CalypsoAI is no longer sold on its own; calypsoai.com redirects to F5's AI Guardrails page. F5 doesn't publish prices for F5 AI Guardrails or F5 AI Red Team. AWS Marketplace shows a $100,000.00 12-month figure for F5 AI Guardrails but says it is a placeholder for a private offer, and lists F5 AI Red Team at $100,000.00 for a 12-month contract. | Nov 19, 2026 |
| TrojAI by A10 Networks | Saint John, Canada | 2019 | 11-50 | No public prices. A10's TrojAI product page directs buyers to contact a product expert or request a demo (checked November 19, 2026). | Nov 19, 2026 |
| Pillar Security | Tel Aviv, Israel | 2023 | 11-50 | No prices on pillar.security (checked November 11, 2026). Vanta says AI Governance by Vanta, which will include Pillar's capabilities, will be a paid offering from January 2027; no price is given. | Nov 11, 2026 |
| Cranium AI | Short Hills, United States | 2023 | — | No plans or prices on cranium.ai; the platform page directs visitors to request a demo (checked November 11, 2026). | Nov 11, 2026 |
| Lasso Security | Tel Aviv, Israel | 2023 | — | No prices on lasso.security; its site index lists no pricing page (checked November 11, 2026). | Nov 11, 2026 |
| ZwillGen AI Division (formerly Luminos.Law) | Washington, DC, United States | 2020 | — | No amounts published. ZwillGen's AI practice page describes monthly retainers, flat or project fees, and hourly rates (checked November 11, 2026). | Nov 11, 2026 |
Buying criteria
What actually matters when picking a red team
- Attack coverage. Does the vendor publish which OWASP LLM Top 10 categories (LLM01 prompt injection through LLM10 unbounded consumption, in the 2025 list) and MITRE ATLAS techniques they exercise? Ask for a public test-suite index, not marketing copy.
- Manual vs automated. Continuous automated red teaming catches drift; expert human red teams find creative bypass chains. Enterprise-grade programs run both. Confirm what percentage of test cases are human-authored.
- Model coverage. Frontier LLMs (GPT-4o, Claude, Gemini, Llama), open weights, RAG pipelines, tool-using agents, image and multimodal, embedded classifiers. If you deploy agents, insist on tool-use adversarial testing.
- Deliverable format. A signed report that maps findings to EU AI Act Article 15, NIST AI RMF Measure 2, and ISO/IEC 42001 clause 8.3 is worth more than raw findings.
- Retest and remediation. One-shot audits are theatre. Confirm scoping for post-fix retesting and a rate card for follow-up rounds.
- Data handling. Your prompts and outputs contain PII, IP, and safety-relevant data. Confirm SOC 2, ISO 27001, data-residency, and whether test payloads are used to train the vendor's own models. Get this in the MSA.
- Independence. For high-risk EU AI Act systems, keep red teaming outside the same provider selling the model or governance platform to avoid conflicts of interest.
Framework mapping
Where red teaming shows up in regulation
| Framework | Requirement |
|---|---|
| EU AI Act | Article 15 mandates appropriate accuracy, robustness, and cybersecurity for high-risk systems. Recital 76 explicitly references adversarial testing. GPAI models with systemic risk face Article 55 obligations including adversarial testing. |
| NIST AI RMF | Measure 2 (2.7 red-team exercises, 2.8 adversarial resilience). NIST AI 600-1 (GAI Profile) explicitly calls for red-team testing across CBRN, cyber, and misuse risks. |
| ISO/IEC 42001 | Clause 8.3 (AI system risk treatment) and Annex A control A.6.2.4 (AI system testing) require documented adversarial testing as part of the AIMS. |
| OWASP LLM Top 10 | Direct threat taxonomy — LLM01 prompt injection, LLM02 insecure output handling, LLM07 system-prompt leakage, LLM08 excessive agency. Most red-team reports index findings to this list. |
| US Executive Order 14110 (rescinded) | Original dual-use foundation model reporting rule was rescinded in January 2025, but the red-teaming reporting cadence it established remains a de-facto industry norm for frontier labs. |
Next steps
Related buying resources
Sources. OWASP LLM Top 10 (2025) · MITRE ATLAS · NIST AI 600-1 GAI Profile · EU AI Act Article 15. Editorial coverage — no paid placement on this page. Contact us to correct any inaccuracy.
Free vendor matchmaking
Ready to run AI red teaming?
Get matched with vendors that run adversarial testing, jailbreak evaluation, and prompt-injection assessments for EU AI Act Article 15, NIST AI RMF Measure 2, and ISO/IEC 42001 clause 8.3.
No sign-up. Independent directory — we do not sell software ourselves.