Popular Security AI tools

33 category leaders in Security, selected from the full directory.

HiddenLayer

  • Runtime and model security against adversarial ML attacks.
  • Defends models, APIs, and AI pipelines in production.
  • Backed by Microsoft M12, IBM, and Capital One.
  • Standout specialist in AI/ML threat defense.

Dropzone

  • Autonomous AI SOC analyst for alert investigations.
  • Clears most Tier-1 noise in minutes, not hours.
  • Connects to existing SIEM, EDR, and ticketing tools.
  • Widely deployed AI SOC copilot for enterprise teams.

Lakera

  • Runtime security for GenAI apps and AI agents.
  • Blocks prompt injection, data leaks, and toxic output.
  • Red teaming plus Gandalf, its widely played jailbreak game.
  • Acquired by Check Point in 2025; brand still ships.

Zenity

  • Leading AI agent security platform.
  • Covers discovery, posture, and runtime controls.
  • Built for enterprise agent governance.
  • Popular with teams securing autonomous workflows.

Promptfoo

  • Open-source LLM eval and red-teaming platform.
  • Catches prompt injection and app vulns in CI.
  • Trusted by Fortune 500 and 300k+ developers.
  • Defining open toolkit for AI application security.

Darktrace

  • Self-learning AI for enterprise threat detection.
  • Autonomous response across network, email, and cloud.
  • Public company; widely deployed in global enterprises.
  • Pioneer in unsupervised ML for cybersecurity.

Abnormal

  • Behavioral AI for enterprise email attack detection.
  • Stops BEC, account takeover, and supply-chain fraud.
  • Integrates with Microsoft 365 and Google Workspace.
  • Leading AI-native email security platform.

Deep Instinct

  • Deep learning malware prevention before execution.
  • Sub-second inference on endpoints and servers.
  • Prevention-first approach with low false positives.
  • Known for zero-day and ransomware blocking.

Hunters

  • Next-gen SIEM with autonomous detection and response.
  • GigaOm Radar Leader; Gartner MQ honorable mention.
  • Pathfinder AI for agentic security investigations.
  • Used by Booking.com, Skyscanner, and ChargePoint.

Vectra

  • AI-driven network detection and response platform.
  • Behavioral analytics for cloud and hybrid attacks.
  • Attack signal intelligence for SOC prioritization.
  • Established NDR leader in enterprise security.

Prisma AIRS

  • AI security platform from Palo Alto (ex Protect AI).
  • Model scanning, red teaming, and runtime controls.
  • Covers apps, agents, and ML supply-chain risk.
  • Enterprise standard for securing AI systems.

Prompt Security

  • GenAI guardrails now part of SentinelOne.
  • Finds shadow AI and blocks prompt abuse.
  • Protects LLM apps, tools, and agent workflows.
  • Category leader for enterprise GenAI security.

Nightfall

  • AI-native DLP across SaaS, endpoints, email, and GenAI tools.
  • Blocks shadow AI exfiltration and governs agent tool calls.
  • Automated remediation with data lineage tracking.
  • Category leader for AI-era data loss prevention.

Exaforce

  • Agentic SOC platform with Exabots for triage and response.
  • $75M Series A; customers include NTT Data and Automation Anywhere.
  • Combines LLMs, semantic models, and statistical ML in one engine.
  • Leading full-lifecycle AI SOC alternative to bolt-on copilots.

Mindgard

  • Automated AI red teaming and runtime agent protection.
  • Maps agent attack surfaces from Lancaster University research.
  • Context-driven guardrails harden production LLM workflows.
  • Standout specialist for continuous AI penetration testing.

Ambient

  • Computer-vision AI for physical security on edge video feeds.
  • Detects threats, safety issues, and ops anomalies in real time.
  • Privacy-preserving alerts for campuses and enterprise sites.
  • Category-defining AI platform for video-based security ops.

Aurascape

  • AI-native platform for shadow AI discovery and runtime controls.
  • AI Proxy secures model traffic; MCP gateway inspects tool calls.
  • $50M funding; founders from Palo Alto Networks and Google.
  • Rising dual-channel leader for enterprise agentic AI security.

Prophet Security

  • Autonomous AI SOC analyst for alert triage.
  • Transparent investigation plans with evidence trails.
  • Integrates with existing SIEM and EDR stacks.
  • Leading agentic SOC platform alongside Dropzone.

Lasso

  • Enterprise-scale secure AI adoption platform.
  • Discovery, governance, and runtime controls for GenAI.
  • Built for regulated teams rolling out agents safely.
  • Category leader for enterprise AI security posture.

Cisco AI Defense

  • Enterprise AI security from Robust Intelligence acquisition.
  • Red teams models and agents; runtime guardrails for MCP.
  • Pairs with Cisco network visibility for AI traffic control.
  • Category-defining hyperscaler AI security suite.

Noma Security

  • Enterprise security for LLMs, RAG, and agentic AI.
  • AI asset discovery, posture, and runtime protections.
  • Built for securing production GenAI deployments.
  • Leading dedicated AI application security platform.

7AI

  • Agentic SOC platform with swarming AI security agents.
  • Automates threat detection, triage, and response workflows.
  • Designed for security operations teams at enterprise scale.
  • Rising AI-native SOC automation brand.

MixMode

  • Third-wave AI for network detection and SOC automation.
  • Unsupervised models adapt to novel attack patterns.
  • Recognized in Gartner NDR competitive landscapes.
  • Strong AI-native cyber defense platform.

Speakeasy

  • AI control plane for agents, MCP servers, and Skills.
  • Extends IdP identity to agents with policy enforcement.
  • Blocks injection, PII leaks, and shadow AI at runtime.
  • Defining governance layer for enterprise AI usage.

Giskard

  • Continuous AI red teaming for LLM and agent apps.
  • Finds hallucinations, injection, and security gaps.
  • Used by L'Oréal, AXA, and Google DeepMind teams.
  • Leading European LLM security testing platform.

Amazon Bedrock Guardrails

  • Managed AWS guardrails for Bedrock GenAI apps.
  • Content filters, topic blocks, and PII redaction.
  • Grounding checks across foundation model traffic.
  • Default hyperscaler controls for safer LLM apps.

Google Model Armor

  • Google Cloud screening for LLM prompts and outputs.
  • Jailbreak, sensitive data, and malicious URL checks.
  • Works with Vertex AI and custom model endpoints.
  • Hyperscaler GenAI safety layer from Google Cloud.

Azure AI Content Safety

  • Azure detection for harmful text and image content.
  • Hate, self-harm, sexual, and violence categories.
  • Prompt shields for jailbreak risk in GenAI apps.
  • Microsoft’s standard content safety API for AI.

NVIDIA NeMo Guardrails

  • Open-source programmable guardrails for LLM apps.
  • Colang policies for safety and dialogue control.
  • Works with NeMo and third-party model stacks.
  • Defining library for production LLM guardrails.

Pillar Security

  • Discovery and governance for enterprise AI agents.
  • Secures agentic workflows across the organization.
  • Policy controls for unsupervised agent actions.
  • Rising leader in agentic workforce security.

Cato AI Security

  • Aim Security's AI firewall, now sold by Cato Networks.
  • Shadow-AI discovery and policy for employee GenAI use.
  • Runtime protection for in-house AI apps and agents.
  • Runs inline on Cato's GPU-powered global SASE backbone.

XBOW

  • Autonomous AI pentester that proves findings with real exploits.
  • First autonomous system to rank #1 on HackerOne (2025).
  • 14,000+ zero-days found in customer applications.
  • Continuous offensive testing for 150+ security teams.

Microsoft Security Copilot

  • Generative AI assistant and agents for security operations.
  • Natural-language incident summaries and threat hunting.
  • Built into Defender, Sentinel, Entra, Intune, and Purview.
  • Agents for phishing triage, alerts, and vulnerability fixes.