AI Security platforms protect organizations from AI-related threats while leveraging AI for enhanced cybersecurity. These tools combine traditional security measures with AI capabilities to detect threats, prevent attacks, and ensure the safe deployment of AI systems.
54 verified AI-first sites in Security.
Cybersecurity threat detection platform. Features autonomous response, network monitoring, and attack prevention. Includes real-time analysis and security automation. Supports enterprise security.
Pricing: Enterprise pricing. Contact for custom solutions.
Network security platform. Features threat detection, attack investigation, and automated response. Includes behavioral analysis and cloud protection. Supports enterprise security.
Pricing: Enterprise pricing. Contact for solutions.
Email security platform. Features behavioral AI analysis, advanced threat detection, and automated response. Includes supply chain fraud prevention and account takeover protection. Supports enterprise email security.
Pricing: Enterprise pricing based on organization size
YC-backed "autonomous security factory": AI security workers that continuously defend repositories, applications, agents, and infrastructure. Integrates with GitHub, Cursor, Claude Code, and Codex. Used by 3,000+ open-source projects; free for open source.
Pricing: Free for open source; paid plans on superagent.sh.
Red Teaming and LLM security platform for detecting hallucinations and security vulnerabilities. Features automated vulnerability detection, prompt injection testing, and continuous red teaming. Trusted by L'Oréal, AXA, BNP Paribas, Société Générale, Decathlon, Michelin, and Google DeepMind. SOC 2 Type II and HIPAA compliant with EU/US data residency options.
Pricing: Open Source: Free. Enterprise: Contact for pricing.
scam detection and prevention platform. Features deepfake detection, AI content analysis, and fraud prevention tools. Includes real-time monitoring and threat detection capabilities. Helps businesses and individuals identify AI-generated scams and fraudulent content.
Pricing: Free tier available. Business plans start at $29/month. Enterprise: Custom pricing.
Purpose-built security platform for AI agents. Threat detection, behavioral analysis, and model integrity checks with post-quantum encryption options. Protects AI systems from classical and quantum-era attacks.
Pricing: Core: $6,250/month. Scaling+: $10,415/month. Quantum: $16,999/month
Physical security AI platform that uses computer vision on edge video feeds to detect threats, safety issues, and operational anomalies in real time. Features automated alerts, privacy-preserving monitoring, and integration with existing security infrastructure for campuses and enterprises.
Pricing: Contact for Pricing
Advanced AI-powered security operations platform featuring swarming AI agents for automated threat response. Features comprehensive EDR investigations, cloud security monitoring, and identity threat detection capabilities. Includes automated alert enrichment, investigation workflows, and security outcome delivery. Particularly valuable for enterprises seeking to augment SOC teams with AI-driven automation.
Pricing: Contact for enterprise pricing
Advanced AI cybersecurity advisor platform combining human expertise with artificial intelligence. Features blockchain-secured audit trails, expert-validated knowledge base, and predictive security prompting capabilities. Includes collaborative analysis tools, multilingual support, and comprehensive API access. Particularly valuable for teams seeking AI-augmented cybersecurity guidance and decision support.
Pricing: Contact for pricing. Enterprise solutions available.
Security operations platform featuring an autonomous AI SOC analyst that automatically triages, investigates, and responds to security alerts. Features agentic AI that autonomously handles threat detection, investigation planning, and evidence collection. Includes transparent investigation workflows showing investigation plans and evidence trails. Offers integration with existing security tools and SIEM platforms. Features automated alert triage and prioritization to reduce SOC team workload. Supports autonomous response capabilities for common security incidents. Includes comprehensive threat investigation tools and evidence analysis. Particularly valuable for security operations teams seeking to automate SOC workflows and reduce alert fatigue. Provides enterprise-grade AI security operations capabilities with transparent, explainable AI decision-making.
Pricing: Contact for pricing.
Deep learning-powered cybersecurity platform specializing in prevention-first malware detection and advanced threat protection. Features deep neural networks trained on billions of files for real-time threat prevention. Includes zero-day malware detection with high accuracy rates. Offers prevention-focused approach stopping threats before execution. Features comprehensive endpoint protection with minimal performance impact. Supports EDR capabilities with behavioral analysis and threat hunting. Includes integration with security operations platforms. Particularly valuable for organizations requiring proactive threat prevention with minimal false positives. Provides enterprise-grade deep learning security with prevention-first approach.
Pricing: Enterprise pricing. Contact for details.
Security platform protecting agentic, generative, and predictive AI applications. Gartner Cool Vendor and Hype Cycle recognized. RSA Innovation Sandbox Most Innovative Startup 2023. Features AIDR (AI Detection & Response) with 100% attack defense at DEF CON/BSidesLV, automated red teaming, model scanning, and supply chain protection. SOC 2 certified. Backed by Microsoft M12, IBM Ventures, Capital One Ventures, Booz Allen.
Pricing: Enterprise pricing. Contact for demo.
Third-wave AI-powered cyber defense platform for threat detection and SOC automation. Gartner 2025 NDR Competitive Landscape recognized. CyberSecurity Breakthrough Award winner 2025. Customers include City of Phoenix, City of Dallas, DOD, US Intelligence Community. Patented self-supervised AI deploys in under 1 hour with 93% alert reduction. Detects zero-day attacks. No GPU required.
Pricing: Enterprise pricing. Contact for demo.
Autonomous AI SOC analyst platform for Tier 1 alert triage and investigation. $37M Series B funding. Gartner Cool Vendor 2024. RSA Innovation Sandbox Finalist. 100+ enterprises including Zapier, UiPath, Lemonade, Fortune 500 companies. Reduces investigation time from 25 minutes to 3 minutes. 90% autonomous Tier 1 ticket clearance. SANS First Look approved.
Pricing: Starting at $36K/year for 4,000 investigations.
Full-lifecycle agentic SOC platform with Exabots for detection, triage, investigation, and response. $75M Series A funding. Multi-model AI engine combining LLMs, semantic models, and statistical ML. Customers include Automation Anywhere, NTT Data, Commonwealth Fusion Systems, LottieFiles. 10x analyst productivity. Available as SaaS or MDR. Reduces MTTA/MTTI/MTTR from days to minutes.
Pricing: Enterprise pricing. Contact for demo.
Next-gen SIEM platform with automated detection, investigation, and response. GigaOm Radar Leader 2025 for SIEM and Autonomous SOC. Gartner Magic Quadrant Honorable Mention 2024. Snowflake Ventures backed. Customers include Booking.com, Skyscanner, ChargePoint, Cimpress, Pennymac. Pathfinder AI for agentic investigations. Pre-built always-on detections by Team Axon.
Pricing: Enterprise pricing. Contact for demo.
Rogue AI Detection System for real-time AI safety monitoring. 24/7 anomaly detection, behavioral analysis, and actionable insights. Prevents unsafe AI behavior across finance, healthcare, legal, tech, retail sectors. 14-day free trial.
Pricing: Free trial (14 days). Contact for pricing.
Enterprise AI security platform for securing LLM, RAG, and agentic AI deployments. Provides AI asset discovery, posture and policy management, and runtime protection for prompts, responses, and tool interactions.
Pricing: Enterprise pricing. Contact for deployment and licensing.
AI-native security platform, part of Check Point since 2025. Protects GenAI apps and agents at runtime against prompt injection, data leaks, and toxic output; governs employee AI use; and offers risk-based AI red teaming. Known for Gandalf, its public jailbreak game.
Pricing: Free tier to start; enterprise plans via Lakera.
F5 AI Guardrails (formerly CalypsoAI): enterprise LLM security and governance with risk controls, policy enforcement, and monitoring for secure generative AI adoption.
Pricing: Enterprise pricing and deployment options; contact sales.
AI control plane for discovering, securing, and governing AI agents, MCP servers, and AI applications across the enterprise. Helps teams inventory agent surfaces, enforce policy, and reduce shadow-AI risk.
Pricing: Free tier; paid plans; see site for pricing.
Guardrails and verification for AI-assisted software development from i3. Structured risk interviews surface data flows, integrations, and failure modes; outputs are versioned guardrails injected into ChatGPT, Gemini, Claude, Copilot, and Cursor workflows. Independent commit-triggered checks, scheduled audits, and on-demand scans with AI-ready fix prompts when constraints drift.
Pricing: Teams from about $150/month; cancel anytime. See site for current tiers.
Composable security APIs for applications, now within the CrowdStrike ecosystem. AI Guard and related services score and filter LLM prompts and responses, redact sensitive data, and enforce recipes for prompt injection and unsafe content. SDKs for Python and common agent frameworks; logging and audit for AI traffic.
Pricing: Free account tier; usage-based and enterprise plans on site.
Continuous AI offensive security for LLM applications, copilots, and agents. Autonomous red teaming that chains exploit techniques, validates findings, and feeds remediation guidance into CI/CD. Black-box through gray-box coverage for prompt abuse, jailbreaks, data exfiltration, and agent manipulation.
Pricing: Enterprise pricing; contact for assessments and platform access.
Agentic application security platform that triages scanner noise and opens production-style fix pull requests. Exploitability-focused analysis across major languages; integrates with GitHub, GitLab, Bitbucket, and Azure DevOps to shrink MTTR for SAST and related findings.
Pricing: Enterprise and team plans; contact for pricing.
AI agent security and governance platform for enterprises: inventory and visibility over agent deployments, configuration and permission posture checks, and runtime monitoring for risky agent actions across common business apps and agent frameworks. Positions AISPM-style controls and detection and response tailored to autonomous agents rather than only static code.
Pricing: Enterprise: contact sales for pricing.
AI agent security and trust platform for evaluating, hardening, and monitoring agent behavior before and after deployment. Provides testing benchmarks, vulnerability detection, and production defense workflows for safer enterprise agent rollouts. Useful for teams reducing model and agent risk in real-world deployments.
Pricing: Free tier available. Team usage-based plans and Enterprise subscriptions available.
AI workflow automation platform for security operations teams, using coordinated agents to triage alerts and accelerate investigation tasks. Helps reduce manual SOC workload through agent-assisted detection, prioritization, and response workflows with human oversight. Useful for organizations modernizing security operations with AI-first automation.
Pricing: Free trial available. Professional and enterprise plans available; see site for current pricing.
AI agent trust platform focused on simulation-driven evaluation, guardrails, and production protection for enterprise agent deployments. Helps teams generate realistic edge-case scenarios, reduce policy violations and hallucinations, and monitor agent reliability with fast, cost-efficient controls.
Pricing: Enterprise pricing; contact for demo and deployment options.
Open-source CLI and library for evaluating, comparing, and red-teaming LLM prompts and applications. Declarative YAML/JSON configs run batch tests across OpenAI, Anthropic, Google, Bedrock, Ollama, and custom providers; supports CI/CD gates, regression checks, vulnerability scanning, and a local web viewer. Test-driven prompt engineering for teams shipping production LLM features. MIT licensed; used in apps serving 10M+ users.
Pricing: Open source. Free.
Agentic AI security control plane for discovery, adversarial testing, and runtime defense. Maps agents across cloud, SaaS, and on-prem; Ascend red-teams behavior; Defend blocks prompt injection and data exfiltration in production.
Pricing: Enterprise pricing; contact for demo and risk assessment.
End-to-end AI security, governance, and agentic platform for enterprises. Discovers and inventories models and vendors, stress-tests deployments, remediates findings, and verifies compliance across the AI supply chain.
Pricing: Enterprise pricing; book demo on site.
Autonomous AI red teaming and runtime protection from Lancaster University research. Maps agent attack surfaces, surfaces exploitable vulnerabilities, and hardens production agents with context-driven guardrails.
Pricing: Enterprise platform pricing; book demo on site.
Enterprise GenAI security for agents and apps across build and deploy stages. AI-BOM discovery, posture management, automated red teaming, and inline runtime enforcement at gateways and proxies.
Pricing: Enterprise pricing; contact for demo.
AI-native DLP preventing sensitive data leaks across SaaS, endpoints, email, and generative AI tools. Blocks shadow AI exfiltration, governs agent tool calls, and automates remediation with lineage tracking.
Pricing: Enterprise subscription; contact for demo and ROI assessment.
Agentic email security for Microsoft 365 and Google Workspace via API-no MX changes. AI agents analyze phishing, BEC, and zero-day threats with analyst-style context, automated remediation, and lower false positives than rule-only gateways.
Pricing: Demo and enterprise plans-see AegisAI.
Palo Alto Networks AI security platform (Protect AI, acquired July 2025). Model scanning, AI-SPM, red teaming, runtime protection, and agent security across the AI lifecycle - from development through production.
Pricing: Enterprise platform pricing; see Palo Alto Networks for Prisma AIRS plans.
Zero-knowledge vault for AI agents: store personal data and context encrypted on-device, then grant one-tap passkey authorization when an agent needs an item. Temporary decrypted access for the task; sensitive values redacted after the session. Works with OpenClaw, Claude, MCP, and similar agents.
Pricing: Free Basic (up to 10 vault items); Pro $9.90/month for unlimited items.
GenAI and agent security platform from SentinelOne. Discovers shadow AI, enforces guardrails, and protects prompts, tools, and LLM apps against injection, data leakage, and abuse - inline with enterprise identity and SOC workflows.
Pricing: Enterprise plans; see prompt.security (SentinelOne) for current pricing.
Self-improving AI SecOps platform: agents for SOC triage, threat hunting, and pentesting that learn from outcomes. Positions as an all-in-one AI security operations layer against AI-driven attacks.
Pricing: See simbian.ai for platform access and plans.
AI-native security platform for shadow AI discovery and runtime controls. AI Proxy secures model traffic; Zero-Bypass MCP Gateway inspects agent tool calls with cross-call lineage. Founded by Palo Alto Networks and Google veterans; $50M funding at 2025 launch.
Pricing: Enterprise platform pricing; contact aurascape.ai.
AI agent security across apps and employee workstations. Workstation Lens shows what Claude Code, Cursor, Codex, Copilot, and browser agents access and run; ARTEMIS handles automated red teaming and ARGUS runtime guardrails, alongside AI asset discovery and agent threat modeling.
Pricing: Enterprise pricing; see repello.ai for demos.
Open-source Apache-2.0 framework for validating LLM inputs and outputs, with validators for PII, toxicity, hallucination, and structured output. Downloaded about 250K times a month. Harvey acquired the team in September 2026; the hosted Hub and remote inference were shut down, so validators now install from PyPI and run locally.
Pricing: Free and open source.
Enterprise AI security platform built on Robust Intelligence. Discovers AI assets, runs algorithmic red teaming and supply-chain scans, and enforces runtime guardrails for LLMs, agents, and MCP traffic. Network-layer options for securing AI usage and development at scale.
Pricing: Enterprise; contact Cisco for AI Defense licensing.
Platform to discover, govern, and secure AI agents and workflows across the enterprise. Inventories the agentic workforce, enforces policy, and reduces risk from unsupervised agent actions. Built for organizations adopting agents at scale.
Pricing: Enterprise plans; contact Pillar Security.
AWS managed guardrails for generative AI on Bedrock. Filters harmful content, blocks denied topics, redacts PII, and applies grounding checks across foundation models. Hyperscaler controls for safer production GenAI apps.
Pricing: Pay-as-you-go with AWS Bedrock; see AWS pricing.
Google Cloud security service that screens LLM prompts and responses for jailbreaks, sensitive data, and malicious URLs. Model-independent controls for Vertex AI and custom endpoints. Enterprise GenAI safety layer from Google.
Pricing: Google Cloud usage-based; see Model Armor pricing.
Microsoft Azure service that detects harmful user-generated and AI-generated content across text and images. Categories for hate, self-harm, sexual, and violence plus prompt shields for jailbreak risk. Integrates with Azure AI apps and Foundry.
Pricing: Azure Cognitive Services pricing; free tier available.
Open-source toolkit for adding programmable guardrails to LLM conversational systems. Colang policies control topicality, safety, and dialogue flows around NVIDIA NeMo and other models. Widely used library for production LLM safety.
Pricing: Open source; see NVIDIA NeMo Guardrails docs.
Cato Networks' AI security product, built from its 2025 acquisition of Aim Security (the Aim brand is retired). Governs employee use of public GenAI, protects in-house AI apps and agents at runtime against prompt injection and data leakage, and adds AI-SPM. Available standalone or as part of the Cato SASE platform.
Pricing: Enterprise plans; contact Cato Networks.
Autonomous SOC platform that investigates alerts from trigger to verdict without playbooks. Deterministic graph orchestration with bounded LLM tasks for enrichment and reporting; humans step in only on decisions that need them.
Pricing: Enterprise plans; see Qevlar pricing.
Autonomous offensive security platform: an AI hacker that finds vulnerabilities and proves each one with a validated exploit. First autonomous system to rank #1 on HackerOne's US leaderboard (June 2025) and on Microsoft's MSRC leaderboard. 14,000+ zero-days found in customer applications; 150+ security teams.
Pricing: Enterprise pricing; see XBOW for plans.
Microsoft's generative AI assistant and agents for security teams. Summarizes incidents, hunts threats in natural language, reverse-engineers scripts, and automates triage across Defender, Sentinel, Entra, Intune, and Purview. Includes agents for phishing triage, alert handling, and vulnerability remediation.
Pricing: Security Compute Units, billed through Azure; included with Microsoft 365 E5.