Menlo's IRIS extension brings security controls inside employees' browsers
Menlo Security's IRIS Browser Extension adds clipboard, upload, screenshot and AI-tool controls inside Chrome, Edge, Safari and Firefox, as corporate AI use jumps from 15% to 45% of employees.

Why it matters
- Menlo Security launched the IRIS Browser Extension on 19 November 2026, running on Chrome, Edge, Safari, Firefox and Chromium-based AI browsers.
- Verizon's 2026 DBIR found regular AI use on corporate devices rose from 15% to 45% of employees in a year, with 67% accessing AI via non-corporate accounts.
- The average company has more than 15% of users running unauthorised AI browser extensions, per Verizon's 2026 report.
- The extension applies controls inside the browser session rather than via proxy routing or traffic decryption, without requiring browser replacement or new hardware.
- CEO Bill Robbins said IRIS 'delivers immediate visibility into where company data is going, including into AI tools.'
Menlo Security launched its IRIS Browser Extension on November 19, 2026, a product designed to give security teams direct oversight of what employees do inside the browsers they already use — from clipboard copies and uploads to the data they paste into AI tools.
The launch lands on a specific, growing problem. Verizon's 2026 Data Breach Investigations Report, cited by Menlo, found that regular AI use on corporate devices rose from 15% to 45% of employees in a single year. The same report found 67% of users accessed AI services through non-corporate accounts on company devices, a figure that puts large volumes of corporate data outside the reach of traditional security controls.
Menlo's argument is straightforward: the browser has become the place where work actually happens, and the tools most companies rely on cannot see it.
Why network and endpoint tools miss the browser
Menlo positions IRIS as filling a gap left by network and endpoint security products. Those tools, the company says, often fail to capture actions that take place within a browser session — copying text, pasting it into a chatbot, uploading a file, printing a page or taking a screenshot.
The shift matters because work has moved into browser-based applications and AI services. Security products built around the network perimeter or the device itself struggle to monitor these routine actions, according to Menlo, and those routine actions have become an important route for data to leave an organisation — particularly when employees use unsanctioned AI tools or personal accounts.
Menlo argues the browser is now the primary exit point for corporate data, and that the industry's answer until now — a dedicated secure browser — asks too much of security teams. IRIS takes the opposite approach: it runs on the browsers employees already use, on Chrome, Edge, Safari, Firefox and Chromium-based AI browsers, and requires no hardware, no browser replacement and no proxy infrastructure.
What does the extension actually control?
The controls operate directly inside the browser session rather than through proxy routing or inline traffic decryption, which lets the extension sit alongside existing network and endpoint products rather than replacing them. Deployed through the Chrome Web Store, Microsoft Intune, JAMF and other mobile device management and unified endpoint management tools, the product's feature set includes:
- Restrictions on clipboard use, printing, uploads and downloads
- Watermarking and read-only pages
- Limits on developer tools
- Measures intended to deter screenshots
- Direction of users toward approved AI tools and web services, with control over the data entered into them
- File inspection for data loss prevention breaches, viruses and malware — with an option to reconstruct infected files so clean versions get delivered rather than blocked outright
- An optional HEAT Shield AI feature for phishing, social engineering and malvertising protection at the moment a user clicks
The file-reconstruction capability is a notable departure from conventional blocking. Instead of stopping a suspicious file at the door, the software rebuilds a clean version and hands that to the user, avoiding the productivity hit of a hard block.
What about the extension problem itself?
User-installed browser extensions are a second target. Menlo said many organisations lack a full inventory of the extensions running across employee devices, even though the risk attached to any given extension can change without warning — an extension that was benign at install time can turn hostile after an update or an ownership change.
Verizon's 2026 DBIR puts a number on the exposure: the average company has more than 15% of users running unauthorised AI browser extensions. IRIS gives security teams an organisation-wide inventory of third-party extensions, with risk scoring and management controls layered on top.
How does it fit compliance requirements?
The extension also ships with audit and compliance functions. Forced authentication to the extension and centralised web logging are designed to support internal audit trails and compliance reviews tied to standards such as SOC 2, HIPAA and GDPR, as well as cyber insurance questionnaires — the paperwork that increasingly demands evidence of data-handling controls rather than assurances.
For regulated industries, that logging layer may matter as much as the real-time controls. Centralised records of browser activity give auditors and insurers a verifiable trail of where company data went and who moved it.
'Security teams know the browser is a blind spot'
Bill Robbins, Chief Executive Officer of Menlo Security, framed the launch as removing the barriers that have kept teams from fixing the problem. "Security teams know the browser is a blind spot. Many have put off fixing it because they didn't want to replace their existing security stack or take on a big migration project. Menlo IRIS removes those obstacles," Robbins said.
"Menlo IRIS slots seamlessly into tools the security team is already running, and even a small team can get essential security running quickly on the browsers employees already use. This delivers immediate visibility into where company data is going, including into AI tools," he said.
The bigger picture
The launch reflects a broader shift in corporate security priorities. As AI assistants, web applications and browser-based workflows absorb more of the working day, the perimeter model — defend the network, harden the device — leaves the highest-traffic channel unmonitored. Menlo's bet is that security teams will accept browser-layer controls only if they arrive without a migration project attached.
With AI use on corporate devices tripling in a year and two-thirds of users reaching AI services through personal accounts, the demand signal is clear. The open question for Menlo and its competitors in the emerging browser-security category is whether an extension-layer approach can match the depth of control offered by dedicated secure browsers — and whether employees will tolerate watermarking, clipboard limits and screenshot deterrence on the browsers they use every day.
Source: itbrief.com.au
More from Elena Vasquez
Show full bio
Market editor covering technology and product news in artificial intelligence and machine learning at AI In Context.
744 articles
Related articles
- Tanium launches AgentSight to track and govern AI tools on endpoints
- Island launches enterprise browser built to control AI agents
- 30% of Workers Fear Autonomous AI Security Risks as Control Shifts
- AI Has Broken Cybersecurity, and Alex Stamos Says a Dark Age Has Begun
- BlackFog survey: 47% report sensitive data exposure via AI