Varonis Introduces Agent Intent-Based Access Control

Rhea-AI Summary

Varonis (Nasdaq: VRNS) introduced Agent Intent-Based Access Control (IBAC), a new Varonis Atlas capability that lets organizations connect AI agents to enterprise data while enforcing safeguards against unsafe or out-of-policy actions.

Agent IBAC evaluates an agent’s reasoning, tool use, and data access against its instructions, with intent-drift detection, full session evaluation, configurable quarantine, human-in-the-loop approvals, and a complete audit trail. It works with existing agents and tools such as Claude Code, Cursor, GitHub Copilot, and Microsoft Copilot Studio and is available now to Varonis Atlas customers.

Market Context

The July 14 Claude Code expansion carried a 6.19% 24-hour reaction, giving this Agent IBAC launch a ...

Analysis

The July 14 Claude Code expansion carried a 6.19% 24-hour reaction, giving this Agent IBAC launch a directly relevant precedent. Current data categorizes short positioning as low; watch for evidence of customer adoption and platform expansion.

Key Figures

Reported rogue-agent cases: one case

1 metrics

Reported rogue-agent cases one case Example involving deletion of an entire production database

Historical Context

5 past events · Latest: Jul 27 (Neutral)

Pattern 5 events

Date	Event	Sentiment	24h Move	Catalyst
Jul 27	Conference participation	Neutral	+2.0%	Planned Black Hat USA and DEF CON appearances featuring AI security demonstrations
Jul 14	AI platform expansion	Positive	+6.2%	Atlas expanded support across Anthropic's Claude enterprise suite
Jul 09	Earnings date announcement	Neutral	-2.8%	Company scheduled its second-quarter 2026 financial results announcement
Jul 08	AI tool integration	Positive	-1.7%	Atlas announced runtime security integration with the Cursor coding tool
Jul 02	Industry recognition	Positive	+2.3%	Varonis received Gartner Customers' Choice recognition for the third consecutive year

24h Move is the share-price change in the day after each event; other market factors may also have contributed.

Pattern Detected

Recent positive AI and product announcements produced mixed reactions, with two aligned events and three divergences.

Key Terms

role-based access control, intent drift detection, multi-turn jailbreak attempts, audit trail

4 terms

role-based access control technical

"role-based access control was never built to judge what a non-human identity"

Role-based access control is a system that gives people access to company systems, files or tools based on the job they do — like issuing keys that only open the rooms needed for each role. For investors, it matters because it reduces the chance of data breaches, fraud, or accidental errors, supports regulatory compliance, and helps preserve business continuity and trust, all of which can affect a company’s financial risk and reputation.

intent drift detection technical

"Intent drift detection: Compares the instruction an agent received"

A monitoring process that detects when the underlying goal, purpose or expected behavior behind user actions, model outputs, or tracked signals changes over time. Like a GPS noticing a driver heading to a different destination, it alerts analysts when queries, customer behavior, or an AI model’s intended task drifts away from what was expected. Investors care because unnoticed intent shifts can make analytics, risk models, fraud detection, or regulatory controls less reliable.

multi-turn jailbreak attempts technical

"such as multi-turn jailbreak attempts"

A series of interactive prompts and replies aimed at getting an AI system to ignore its built-in safety rules or reveal restricted content. Like someone asking a guard increasingly specific questions and rephrasing them after each response until the guard lets something through, these repeated exchanges exploit context and memory across turns to circumvent safeguards. For investors, the term signals a type of operational and reputational risk tied to AI products and disclosures.

audit trail regulatory

"Complete audit trail: Records every prompt, response, and tool execution"

A chronological record that shows who performed each action, when it happened, and what was changed across financial systems and documents; it logs transactions, edits, approvals, and access so every step can be traced. For investors, an audit trail matters because it boosts trust and accountability—helping detect errors or fraud, supporting audits and regulatory checks, and making financial reporting and due diligence clearer, like a transaction-level receipt for a company’s records.

AI-generated analysis. How Rhea-AI works. Not financial advice.

See more from StockTitan in Google Search and AI answers. Adds StockTitan as a preferred source · opens Google

Add on Google Not now

New Varonis Atlas capability lets businesses connect AI to their data while detecting and stopping agents that drift from their purpose

MIAMI, Aug. 03, 2026 (GLOBE NEWSWIRE) -- Varonis Systems, Inc. (Nasdaq: VRNS), the data and AI security leader, today announced Agent Intent-Based Access Control (IBAC), a new capability in Varonis Atlas that lets businesses connect AI agents to their enterprise data with safeguards that stop dangerous or out-of-policy behavior.

Agents are making headlines for going rogue, exposing sensitive company data and, in one case, deleting an entire production database.

Agents need broad access to data and tools to be useful, which is precisely what makes them risky, and role-based access control was never built to judge what a non-human identity does with the access it has.

"The question is no longer ‘Can a user access this data?’ but ‘In this context, should this agent be allowed to take action on this data?’" said Ron Bennatan, VP of AI and Data Security Strategy at Varonis. "We built Agent IBAC to give organizations assurance that their agents are acting as intended and not putting their business at risk."

How Agent IBAC Works

Agent IBAC checks whether an agent’s reasoning, tool use, and data access are consistent with its assigned instructions, and alerts on or blocks actions that do not align.

When an agent crosses the line, Atlas can quarantine the identity behind it and block everything that follows for a defined window.

Key Capabilities:

• Intent drift detection: Compares the instruction an agent received to its reasoning, tool calls, and data access with lenient, balanced, and strict sensitivity settings.
• Full session evaluation: Reviews every prompt, response, and tool call in a session to catch risk that builds gradually, such as multi-turn jailbreak attempts. Customers can also define their own session policies in plain language.
• Quarantine: Blocks an identity or session after a violation for a window the customer sets, with admin controls to lift, extend, or make it permanent.
• Human in the loop: Routes a flagged action to a person for approval rather than blocking it outright.
• Complete audit trail: Records every prompt, response, and tool execution alongside the action Atlas took, for security, governance, and compliance teams.
  
  

Agent IBAC works with the agents and AI tools organizations already run, including Claude Code, Cursor, GitHub Copilot, and Microsoft Copilot Studio. It is available today to Varonis Atlas customers and is part of the platform's broader approach to securing agents from code to runtime.

Additional Resources: 

• Learn more about Agent IBAC.
• Request your Varonis Atlas demo. 
• Visit our blog, and join the conversation on LinkedIn and YouTube. 
  
  

About Varonis 
Varonis (Nasdaq: VRNS) secures AI and the data that powers it. The Varonis platform gives organizations automated visibility and control over their critical data wherever it lives and helps ensure safe and trustworthy AI from code to runtime. Backed by 24x7x365 managed detection and response, Varonis gives thousands of organizations worldwide the confidence to adopt AI, reduce data exposure, and stop AI-powered threats. 

Investor Relations Contact: 
Tim Perz 
Varonis Systems, Inc. 
646-640-2112 
investors@varonis.com

News Media Contact: 
Rachel Hunt 
Varonis Systems, Inc. 
877-292-8767 (ext. 1598) 
pr@varonis.com

FAQ

What is Varonis Agent Intent-Based Access Control (IBAC) for VRNS?

Varonis Agent Intent-Based Access Control (IBAC) is a Varonis Atlas capability that governs AI agents’ access to enterprise data. According to Varonis, it monitors agents’ reasoning, tool use, and data access to detect intent drift and block or flag out-of-policy behavior.

How does Varonis Agent IBAC work to control AI agents’ behavior?

Agent IBAC compares each agent’s instructions with its reasoning, tool calls, and data access, then intervenes on misaligned actions. According to Varonis, it applies session-wide evaluation, sensitivity settings, quarantines, human-in-the-loop approvals, and audit trails to manage risky or out-of-policy AI behavior.

Which AI tools and agents are supported by Varonis Agent IBAC (VRNS)?

Varonis Agent IBAC works with AI agents and tools organizations already run, including several popular coding and copilot systems. According to Varonis, supported tools include Claude Code, Cursor, GitHub Copilot, and Microsoft Copilot Studio, integrating with existing AI deployments.

How does Varonis Agent IBAC help prevent rogue AI agents from exposing sensitive data?

Agent IBAC detects when an AI agent’s actions drift from its assigned intent and can block violations. According to Varonis, it reviews prompts, responses, and tool calls, supports quarantine windows, and offers human review to stop unsafe access that could expose sensitive company data.

Is Varonis Agent Intent-Based Access Control available now and to whom?

Varonis Agent Intent-Based Access Control is available immediately to customers using the Varonis Atlas platform. According to Varonis, Agent IBAC is delivered as part of Atlas’s broader approach to securing AI agents from code through runtime across enterprise environments.

What are the key features of Varonis Agent IBAC for securing enterprise AI?

Varonis Agent IBAC offers intent drift detection, full session evaluation, configurable quarantine, human-in-the-loop approvals, and comprehensive audit trails. According to Varonis, these features help security, governance, and compliance teams control AI agents’ data access and document Atlas’s responses to risky actions.