Enterprise AI security vendors shape emerging M&A map for agent controls

Traders Union
All News
Financial News
Enterprise AI Security for Agents

Highlights

AI agents now represent a distinct enterprise identity class, requiring companies to track agent access, system connections, and authorization levels.
The AI security market is fragmenting around specialized control points such as agent identity, data controls, prompts, and auditability rather than forming a single broad category.
Recent transactions include Kiteworks' acquisition of Bonfy.AI for real-time data classification and Huskeys' $27 million Series A led by Blackstone to secure advanced internet traffic.

Agent identity becomes a new security layer

As reported by Crunchbase News, AI agents are emerging as a distinct class of enterprise identity because they can access files, query databases, send emails and execute code inside corporate environments. Once agents receive that level of access, companies need to track what information they use, which systems they connect to and whether their actions are authorized.

As deployments expand from small tests to potentially hundreds of agents, security teams are facing a more complex control problem than with traditional users or service accounts. Unlike passive identities, agents can move across systems, invoke tools and make decisions, increasing the need for tighter governance and monitoring.

Specialized control points drive deal activity

The article argues that the market is unlikely to develop as one broad AI security category and instead will form around narrower control points. Those areas may include agent identity, data access controls, prompts, MCP servers, plug-ins, network traffic and auditability.

Recent activity already reflects that segmentation. Kiteworks acquired Israeli startup Bonfy.AI, which focuses on real-time data classification and policy enforcement, while Israeli cybersecurity startup Huskeys raised a $27 million Series A led by Blackstone to secure increasingly complex internet traffic, including traffic generated by autonomous systems.

That fragmentation is beginning to shape a new M&A map for cybersecurity, cloud and enterprise software vendors that may need agent-security capabilities inside their products. For founders, the implication is that a broad AI security label may be less valuable than a clear position around the specific control layer their product manages.

In our earlier coverage of the OECD’s updated outlook, we noted that heavy investment in AI infrastructure—such as data centers and semiconductors—was helping cushion global growth in 2026 despite an energy-price shock tied to the Middle East conflict. The OECD also warned that inflation pressures and volatility risks could intensify in 2027, especially if energy turbulence and weaker-than-expected returns on AI investment coincide.

This material may contain third-party opinions, none of the data and information on this webpage constitutes investment advice according to our Disclaimer. While we adhere to strict Editorial Integrity, this post may contain references to products from our partners.