>
All Issues

AI Security Weekly

Issue #1 — March 2026

Real-time intelligence for the enterprise AI security market

Published March 13, 2026 12 min read 5 Sections

Top Market Developments

01

Gartner Projects $244B in Security Spending for 2026

Enterprise security spending is on track to reach approximately $244 billion in 2026, roughly $29 billion more than 2025. AI security stands out as the fastest-growing segment, with the AI-amplified security market projected to reach $160 billion by 2029, representing a 3.3× increase from its $49 billion base in 2025.1 This surge reflects the enterprise recognition that securing AI systems is no longer optional — it is a board-level priority driving budget allocation across industries.

02

Consolidation Wave Accelerates

A historic wave of acquisitions reshaped the AI security vendor landscape throughout 2025 and into 2026. Palo Alto Networks acquired Protect AI for approximately $500 million (closed July 2025) and subsequently launched Prisma AIRS 2.0. Check Point acquired Lakera for roughly $300 million (closed Q4 2025). F5 acquired CalypsoAI for $180 million (October 2025). SentinelOne announced the acquisition of Prompt Security for approximately $250 million (August 2025). Apple quietly acquired WhyLabs.2,3,4,5 The message is clear: standalone AI security is being absorbed into platform plays at scale.

03

EU AI Act High-Risk Enforcement Arrives August 2026

The most consequential provisions of the EU AI Act take effect on August 2, 2026, requiring high-risk AI systems to demonstrate compliance with risk management, data governance, transparency, and cybersecurity measures. Penalties can reach up to €35 million or 7% of global annual revenue, whichever is higher.6 Meanwhile, multiple US states — including California, Colorado, Texas, and Illinois — are enacting their own AI governance laws in 2026, creating a complex patchwork of compliance obligations for enterprises operating across jurisdictions.

04

AI Agent Security Crisis Escalates

Gravitee's State of AI Agent Security 2026 report reveals a stark gap between adoption and security readiness. While 81% of teams have moved past planning into production with AI agents, only 14.4% have received full security approval. An alarming 88% of organizations have reported confirmed or suspected AI agent security incidents. Only 47.1% of deployed AI agents are actively monitored, leaving a majority of production agents operating without meaningful security oversight.7

Vendor Spotlight

Fiddler AI

Spotlight
CEO Krishna Gade
Latest Round $30M Series C
Lead Investor RPS Ventures
Total Funding $100M
Date Jan 27, 2026
Recognition #1 AI Agent Security, CB Insights 2025

Fiddler AI is positioning itself as the "first control plane for AI," providing a unified layer of standardized telemetry, evaluation, continuous monitoring, enforceable policy, and auditable governance across the entire AI lifecycle. The company's platform addresses a critical enterprise need: as organizations deploy dozens or hundreds of AI models and agents, they require a single pane of glass for observability, compliance, and risk management. Fiddler serves regulated industries including the US Navy.8

Why It Matters

Fiddler's "control plane" framing signals a maturing market that is moving beyond point solutions for prompt injection or model scanning toward comprehensive AI governance platforms. Watch for Fiddler to compete directly with the AI governance capabilities now being assembled by Palo Alto Networks and Check Point through their recent acquisitions.

The Rise of AI Agent Security

80%

of organizations encountered risky AI agent behaviors9

67%

already using agentic AI in production10

AI agents do not fit existing security models. They are "hyper-scale, dynamic, and short-lived entities" requiring entirely new approaches to identity and privilege management. The Cloud Security Alliance predicts 2026 as the defining year for new agent security benchmarks, with the MAESTRO framework emerging as a key governance model.11

Vendor Responses

Credo AI — Agent Registry for cataloging and governing AI agents
Arthur AI — Agent Discovery & Governance for enterprise oversight
Fiddler AI — Agentic observability and lifecycle monitoring
HiddenLayer — AI Discovery for detecting shadow AI agents

Enterprise Buyer Signal

83% plan to deploy agentic AI

Only 29% feel ready to do so securely

Cisco State of AI Security 202612

90%+

of organizations do not allow blanket AI access; 56% actively blocking most AI tools

16,200

AI-related security incidents in 2025 — a 49% increase year-over-year13

50%+

of incidents concentrated in finance and healthcare sectors

New Vendor Watchlist

01

Cisco Foundation AI

Built from the Robust Intelligence acquisition team. Open-source approach with an 8B parameter model outperforming larger LLMs on security tasks.14

02

Palo Alto Networks Prisma AIRS 2.0

Protect AI integration complete. End-to-end AI agent security, shadow AI detection, and Google Cloud partnership.3

03

Check Point AI Security Platform

Post-Lakera acquisition. Self-described "world's first end-to-end AI Security Platform" covering the full AI lifecycle.4

04

Asenion (formerly Fairly AI)

Acquired anch.AI, integrated with IBM watsonx.ai. Targeting EU AI Act compliance as primary market wedge.15

Subscribe for Weekly Intelligence

Get every signal, every shift, and every decision point — distilled into one actionable briefing, delivered weekly.

Subscribe Now →

Sources & Citations

Gartner, "Forecast: Information Security, Worldwide, 2023–2029" (2025); Statista, "AI in cybersecurity market forecast 2025–2030."

SentinelOne, "SentinelOne to Acquire Prompt Security" press release (August 2025).

Palo Alto Networks, "Palo Alto Networks Completes Acquisition of Protect AI" (July 2025); Prisma AIRS 2.0 launch announcement.

Check Point Software, "Check Point Acquires Lakera to Strengthen AI Security Platform" (Q4 2025).

F5 Networks, "F5 Acquires CalypsoAI" (October 2025); reporting on Apple's acquisition of WhyLabs.

European Commission, Regulation (EU) 2024/1689 — EU AI Act, Article 99 (penalties), Article 113 (timeline).

Gravitee, "State of AI Agent Security 2026" report (2026).

Fiddler AI press release, "$30M Series C" (January 27, 2026); CB Insights, AI 100 2025.

McKinsey & Company, "The state of AI: How organizations are rewiring to capture value" (2025–2026).

Kiteworks, "2026 AI Security Report" (2026).

Cloud Security Alliance (CSA), MAESTRO Framework for AI Agent Security (2026).

Cisco, "State of AI Security 2026" report.

Industry aggregate data; multiple sources including Cisco AI Security Report and Gravitee report.

Cisco, "Foundation AI: Robust Intelligence Team Launches Open-Source Security Model" (2026).

Asenion (formerly Fairly AI), company announcements re: anch.AI acquisition and IBM watsonx.ai integration.