Vendors with new products and capabilities that uncover unauthorized “shadow AI” tools and secure AI and agent usage in real-time have been named to CRN’s 2026 AI 100 list.
From protecting sensitive data in AI applications to protecting the activities of AI agents, a huge number of vendors in the cybersecurity industry are moving into the AI security space. The proliferation of all kinds of AI-powered tools for productivity has created a high demand for new products and features from security vendors, especially those that can reduce the risk of data breaches and other cyber-attacks via the AI attack surface.
[Related: 10 Key AI Security Controls For 2026]
As part of CRN’s 2026 AI 100, we selected 20 cybersecurity vendors that have introduced new capabilities focused on increasing visibility through detection into both how employees are using sanctioned AI tools and the use of unsanctioned “shadow AI” systems.
Another major focus is the rise of autonomous or semi-autonomous AI agents and the need to secure their activities in real-time to prevent data breaches and other cyber risks. In this sense, security and management around identity, access, and privileges are critical, as reflected in many of the new products currently on the market.
At the heart of the latest AI security product releases is a focus on providing real-time enforcement and detecting threats as they occur, not just after the fact. Ultimately, the urgency of concerns around AI and agent AI is driving significant investment in new capabilities in the security industry, creating significant new opportunities for solution providers working with leading vendors.
Below are the 20 hottest AI security companies in 2026.
1 password
david forgno
CEO
1Password recently debuted a new unified access platform aimed at improving discovery, security, and audit access for all identities, including AI agent identities. The Toronto, Ontario-based company says its new 1Password Unified Access product spans human, machine, and AI agent identities, providing a clear model for providing access security in the AI space.
Kato Networks
Shlomo Kramer
Co-founder, CEO
Cato Networks expanded its SASE Cloud platform to secure AI deployments with the acquisition of Aim Security in September 2025. The acquisition of the Tel Aviv, Israel-based company brings capabilities to secure the use of both public and private AI applications, including AI agents and AI Security Posture Management (AI-SPM).
Check Point Software Technologies
Nadav Zafrir
CEO
Check Point Software Technologies, based in Tel Aviv, Israel, recently announced its new AI Defense Plane product, aimed at providing enhanced security for agent-enabled enterprises. AI Defense Plane provides key modules including employee AI security, providing visibility, governance, and runtime protection for employee use of AI.
cloudflare
Matthew Prince
Co-founder, CEO
Cloudflare recently expanded its Cloudflare One platform with new features such as AI Security Posture Management (AI-SPM) to protect and enforce policies for the use of AI. The San Francisco-based company said key features include discovering how employees are using AI, protecting against unauthorized “shadow AI” use, and protecting sensitive data.
cloud strike
George Kurtz
Co-founder, CEO
CrowdStrike recently announced the general availability of its Falcon AI Detection and Response product, which enhances security around AI prompts and agent interactions, as well as AI agent detection and shadow AI governance. The Austin, Texas-based cybersecurity giant also announced a new Falcon Data Security product featuring protection for sensitive GenAI data.
sierra
Yotam Segev
Co-founder, CEO
Cyera recently released a series of new features to protect AI deployments, including a new Browser Shield tool. New York-based Cyera says the product provides real-time visibility and discovery of AI used within the browser, regardless of whether those tools are managed or unmanaged.
dark trace
ed jennings
Representative Director and President
Darktrace recently announced the release of its new Adaptive Human Defense product, which provides an AI-era alternative to traditional security awareness training. The Cambridge, UK-based company said Adaptive Human Defense provides real-time adaptive coaching to employees, leveraging behavioral AI to educate users on the job.
fortinet
Ken Shee
Co-founder, Chairman, CEO
Fortinet’s FortiAI products provide AI security across infrastructure, models and workloads, as well as data and supply chains, according to the Sunnyvale, Calif.-based company. Key features include LLM security and real-time scanning to prevent unauthorized “shadow AI” use, the company said.
netscope
Sanjay Beri
Co-founder, CEO
Netskope recently debuted a new suite of AI security products within the Netskope One platform, including a new Agentic Broker that provides visibility and control of MCP transactions. Other new products from the Santa Clara, Calif.-based company include inspection and enforcement of security policies related to private AI applications and LLMs, as well as AI-specific threat prevention.
Octa
tod mckinnon
Co-founder, CEO
Okta recently published what it calls a “New Blueprint for a Secure Agent Enterprise,” revealing a new framework to address the most important issues in agent AI deployments. Implementation of the framework will be enabled in part through the Okta for AI Agents product, which will be available in April, the San Francisco-based company said.
orca security
Gil Geron
Co-founder, CEO
Orca Security recently announced the addition of new AI security capabilities, including runtime AI threat detection. Orca, based in Tel Aviv, Israel, said the update enables it to identify all interactions between workloads, identities, and other processes and AI systems, allowing it to understand how AI is being used while detecting data breaches.
palo alto networks
Nikesh Arora
Chairman, CEO
Palo Alto Networks recently announced Prisma AIRS 3.0, its next-generation AI security platform. It provides discovery capabilities such as instant inventory of AI agents and models and connections between their environments. The Santa Clara, California-based company says Prisma AIRS 3.0 maps the complete architecture of an AI agent while also providing a means to scan for vulnerabilities within the agent.
proof point
Sumit Dhawan
CEO
Proofpoint announced that it is extending its AI-powered data security posture management (DSPM) capabilities to on-premises environments as a way to protect a company’s entire IT deployment. The Sunnyvale, Calif.-based company said the move will help its many partners and customers who are now enabling AI adoption, allowing teams to provide data access at the appropriate privilege level regardless of location.
rubric
Vipul Sinha
Co-founder, CEO
Rubrik recently announced the launch of the “first AI governance engine” in the data security industry that can provide real-time security and control to autonomous agents. The Palo Alto, California-based vendor said its Semantic AI Governance Engine includes key features such as semantic policy interpretation to translate natural language instructions into machine logic.
sail point
mark mcclain
Founder, CEO
SailPoint recently announced the debut of Shadow AI Remediation, which provides detection, monitoring, and security for the use of unauthorized “shadow AI” tools. The Austin, Texas-based company says key capabilities include real-time visibility into unmonitored AI tools and proactive remediation of unauthorized tool abuse.
sentinel one
Tomer Weingarten
Co-founder, CEO
Recent major announcements from SentinelOne include a new Prompt Security product that protects the use of AI within on-premises environments, including sovereign and air-gapped data centers. The Prompt Security On-Premise product provides self-hosted AI security with detection of unauthorized “shadow AI” use and real-time redaction of sensitive data, the Mountain View, Calif.-based company said.
trend AI
Eva Chen
Co-founder, CEO
TrendAI recently announced the debut of the TrendAI Agentic Governance Gateway, which provides visibility and control of interactions between autonomous AI agents. The Irving, Texas-based company said key capabilities include understanding the intent behind communications between agents, visibility across the systems agents interact with, and enforcing policies on agent actions.
upwind
Amiram Shachar
Co-founder, CEO
Upwind recently announced its AI security product portfolio, which includes AI-SPM (AI Security Posture Management), AIDR (AI Detection and Response), and AI-BOM (AI Bill of Materials). The Tel Aviv, Israel-based startup also debuted AI network visualization, MCP security, and AI security testing capabilities.
With
Asaf Rappaport
Co-founder, CEO
Wiz recently announced the launch of a new AI application protection platform aimed at providing “end-to-end” security for AI applications, the Google-owned company said. New York-based Wiz said the platform will provide developer and security teams with the necessary context to understand real-world cyber risks and enable those teams to strengthen their security posture.
Z scaler
Jay Chowdhury
Founder, Chairman, CEO
Zscaler recently announced the new Zscaler AI Security Portfolio with features such as AI Asset Management and Secure AI Access. The San Jose, California-based company said its AI security suite also includes features to protect AI infrastructure and applications, including AI red teaming, prompt hardening, and runtime guardrails.
