- Moriya Pedael of GeoEdge told Black Hat that the ad chain lets many unknown parties add code to one ad, and publishers often cannot control it.
- Her LANJack case showed an ad-borne attack scanning internal networks for IoT devices. She advised owners to secure their devices, and said she does not know whether harm followed.
Moriya Pedael, a security researcher at GeoEdge, argued that online advertising has become a way to run other people's code inside trusted places. That includes the network behind a visitor's router. She made the case in a Black Hat USA 2026 session called "LANJack: Turning Ads into IoT Recon Tools."
What was said
Pedael described how ads used to work. Publisher and advertiser dealt directly, and the publisher knew which ad it was showing. That was too manual, so new middlemen entered the chain. Now, she said, publishers cannot steer which ads appear on their pages.
Each party in the chain contributes code, so the finished ad is a bundle of third-party code. Publishers can limit what ad code does with tools such as CSP and sandboxing. Pedael said they often skip this, since it can cost them ad revenue and features. Her conclusion: "the ad ecosystem actually became a distributed platform for executing third-party code inside trusted environments."
LANJack was her example. The session notes describe it as the first known campaign of its size to use a trick called DNS rebinding, spread through mainstream ad-buying systems. In DNS rebinding, an attacker's domain first points to a harmless address. A second lookup then points the same name to an address inside the victim's network. The browser still sees the same site name, so it treats the connection as safe. Pedael said the campaign used this to scan for IoT devices, meaning connected gadgets such as cameras and routers.
Why it matters
Our reading: this moves ad risk out of the marketing budget and into network security. The people who buy and place ads do not run the cameras, routers and other devices on a home or office network. The people who run those devices do not choose the ad partners. No single job covers that gap.
Pedael did not name an accountable party. She did say publishers do not consistently use the controls they have. Her closing advice also put the work on device owners: protect devices with secure communication and authentication. For managers, the practical question is who in your company can see which third-party code runs on your pages and your visitors' networks.
The other side
Pedael was open about what she could not show. She said the campaign looked like the first, reconnaissance phase of an attack. She did not know how much information the attacker got out, or whether a more specific attack followed.
She also did not say that one broken control caused this. Her view was that the attacker combined several pieces: trust placed in ads, the way browsers behave, and side channels from defensive tools like CSP. That makes the problem harder to assign to one party.
Her own advice is modest: "if the browser can access a resource, don't assume that the attacker can." It is a design rule, not a fix for the ad chain itself.
Written by the WebPulse Newsroom with AI assistance, and checked by our editorial review: every quotation was verified against the recording's transcript. How we use AI.
The conversation this talking point comes from
- Black Hat: Black Hat USA 2026 | LANJack: Turning Ads into IoT Recon Tools (2026-11-16)





