Bay Raises $50 Million to Secure Enterprise AI Agents with Runtime Control Plane

Tel Aviv-based Bay has emerged from stealth after raising $50 million in funding, including a $37 million Series A led by Battery Ventures and a $13 million seed round led by YL Ventures and Striker Venture Partners. Founded in February 2026 by Yuval Zacharia, Boris Danilovich, and Aner Morag, all alumni of Israeli Military Intelligence Unit 8200, the company addresses the challenge of unpredictable AI agent behavior in enterprise environments. Bay offers a single control plane that automatically discovers agents, coding assistants, MCP servers, skills, extensions, and local models while mapping their connections and risky tool combinations. Its Enforcer component evaluates every proposed action in the context of the full agent session, deciding in milliseconds whether to allow, refer, redirect, or deny the action based on policies generated by the embedded Lighthouse AI system. The funding will support product development, engineering expansion, and go-to-market efforts in the US and Europe, with the long-term goal of becoming the core control layer for organizations deploying autonomous AI agents at scale.

SecurityWeekNovember 11, 2026Translated from English

Read at source

Bay, a Tel Aviv-based startup, has emerged from stealth with $50 million in total funding to help enterprises safely deploy and manage AI agents. The round comprises a $37 million Series A led by Battery Ventures and a previously undisclosed $13 million seed round led by YL Ventures and Striker Venture Partners.

The company was founded in February 2026 by Yuval Zacharia (CEO), Boris Danilovich (VP of R&D), and Aner Morag (CPO). All three founders are alumni of the IDF Unit 8200, Israel’s elite military intelligence unit. Bay’s platform is designed to give organizations visibility and control over agentic AI systems whose behavior is determined at runtime based on user permissions and prompts.

When deployed, Bay automatically builds a complete inventory of AI agents, coding assistants, MCP servers, skills, extensions, and local models in use. It maps how these components connect and identifies potentially risky combinations of tools. The solution follows each agent session from the initial prompt through to the final action.

The platform’s Enforcer component evaluates every proposed action in the full session context and decides within milliseconds whether to allow it, refer it to a human, redirect it according to policy, or deny it outright. Policies are automatically generated and maintained by Bay’s embedded AI system called Lighthouse, which learns from the organization’s own historical activity.

CEO Yuval Zacharia emphasized three core design principles: effortless deployment, immediate time to value, and a clear path to operationalization. These led to an agentless architecture deployable across an entire fleet in under five minutes and patented retrospective analysis of 30 days of historical logs to demonstrate value quickly.

The company plans to use the new capital to further develop the product, expand its engineering team, and invest in go-to-market initiatives across the United States and Europe. Longer term, Bay aims to serve as the central control layer for enterprises running large-scale, autonomous AI agents while expanding the scope of tasks agents can safely perform.

Bay Battery Ventures Yuval Zacharia IDF Unit 8200 Lighthouse

ShareX Facebook LinkedIn Reddit

Meridian digestThe week's most consequential security stories, distilled into one briefing.See the digest

More in AI & Model Risk

AI & Model Risk

Token's Rehearsal Agent Breaks Core Security Rules on Identity, Access, and Traceability

Token built an AI browser agent named Rehearsal that deliberately violates three fundamental security principles for AI agents: separate identities, least-privilege access, and independent traceability. The agent runs inside a user's authenticated browser session, making its actions indistinguishable from human clicks to backend applications. Because Token controls both the agent and the platform, the company implemented controls such as read-only dashboards, session recording, and explicit user approval for changes. The experiment revealed that external agents lack these safeguards and highlighted gaps in emerging standards including WebMCP and Web Bot Auth. Security teams are advised to begin inventorying agentic browsers, mapping delegated access, and enforcing step-up authentication for sensitive actions until holistic standards mature.

The Hacker NewsNov 11, 2026Translated from English

AI & Model Risk

Google Expands Gemini 4 Carbon Access After AI Agents Formally Verify Signal Protocol Code

Google has broadened access to its Gemini 4 Carbon model, allowing paying developers, businesses, and Google AI Ultra subscribers to use it through the Gemini API and Gemini for business. The expansion follows the successful use of Carbon-based AI agents to formally verify core security properties of the Signal protocol implementation in libsignal. Google DeepMind also introduced AlphaProof Code, which combines human experts with AI agents to generate millions of lines of Lean proofs that confirm guarantees such as forward secrecy and post-compromise security at the machine-code level. The verification effort, completed in four weeks by five researchers, also demonstrated the method's ability to detect previously patched flaws when they were reintroduced. In parallel, Carbon for cyber helped Wiz discover zero-days at two major SaaS providers and assisted Google's Cloud Vulnerability Research team in confirming the exploitability of two bugs in a popular open-source project. Google is releasing the specifications and proofs publicly while noting that the work applies to its own build of the code rather than Signal's production applications.

SecurityWeekNov 10, 2026Translated from English

AI & Model Risk

CISO Forum Virtual Summit 2026 Opens Today with Focus on AI Governance and Enterprise Defense

The CISO Forum Virtual Summit 2026 runs November 10-11, 2026, from 11:00 AM to 4:00 PM ET each day, offering senior cybersecurity leaders two days of sessions on AI-driven threats and risk management. The event features practical frameworks for AI security programs, runtime identity protection, supply chain defense, and team resilience. Day One includes talks on securing GenAI within 30-60-90 days, evolving AI threat vectors, machine identity security, evidence-based risk metrics, and mitigating burnout as a security risk. Day Two covers AI agents in production, accountability frameworks, supply chain vulnerabilities amplified by AI, and lessons from high-stakes environments such as Kyiv’s municipal cybersecurity operations. Speakers represent organizations including F5, Censys, ClearVector, Horizon3, Snyk, Sonatype, and BreachRx. Attendees can register at virtual.cisoforum.com for live and on-demand access.

SecurityWeekNov 10, 2026Translated from English