Domain abuse used to be mostly opportunistic. Someone registers your brand name in a new top-level domain, waits for you to notice, and either uses it for fraud or tries to sell it back at a premium. Frustrating, but predictable enough to manage. Our partners at Brand Safety Alliance (BSA) document a new, entirely different wave of AI domain abuse in their Domain Protection Report 2026. This read is for brand protection teams that still rely on slow, human review processes.
How AI has changed domain abuse
The Chair of Internet, Domain Name, eCommerce and Social Media Practice at Greenberg Traurig, puts it plainly in the report: “Bad actors have become way more sophisticated. They’re leveraging free tools, and agentic AI is driving their costs down to almost nothing.”
AI agents can register thousands of domains simultaneously, swap brand names across templates, and launch coordinated phishing campaigns around the clock without meaningful human involvement.
A brand protection strategist at Stobbs, adds a forward-looking prediction, “The rapid emergence of new technologies has driven huge increases in scam volumes, and will continue to do so. Generative AI has dropped the technical entry barrier for bad actors, and has also created the potential for entirely new forms of infringement.”
The prediction is spot on. We’re already seeing tens of thousands of domains registered with keywords like “ai” or “crypto,” and ICANN’s new gTLD application round will only accelerate this further.
Why detection alone is no longer enough
One of the themes in the BSA report is the limitation of AI-based detection. The Senior Legal Director at Uber, describes the challenge: “We’ve seen ‘obfuscation,’ where counterfeit merchandise sellers will make adjustments to the images they’re using by pixelating them, blurring them, doing things to cover our logo. To an AI detection system, it is different enough. It doesn’t hit the similarity score they’re looking for, so it’ll pass on this thing. But we can find them doing standard word searches.”
The same AI technology that empowers bad actors also creates blind spots in the tools used to catch them. We recently wrote about the need for human intelligence in monitoring, not just AI, and Uber has come to the same conclusion. They use a hybrid approach that combines AI image search with manual methods because neither works reliably on its own.
What a modern domain protection strategy actually looks like
Greenberg Traurig’s advice in the report reframes the goal: “The motivation for abuse is purely about money. Scammers are running a business like anyone else. They have accountants, and they care about ROI.”
That reframe matters. Brand defense is risk mitigation. The goal isn’t to catch every abusive domain. It’s to make your brand an unprofitable target.
That means shifting the strategy from reactive enforcement to proactive prevention, and from siloed IP/legal functions to what the report calls a whole-of-organization approach. Legal, marketing, security, and finance all have a stake in how AI domain abuse affects brand reputation, customer trust, and litigation costs. Keeping brand protection isolated from those teams limits both the resources and the response.
Our recommendation for brand teams
- Monitor domain registrations proactively. Our trained experts know how to look for patterns that indicate emerging threats.
- Reduce enforcement reliance. Invest in proactive protection measures that prevent registrations rather than requiring you to recover domains after the fact.
- Plan ahead for new TLDs. The next expansion of the domain namespace is already underway. Brands that don’t have a strategy for it will spend the next few years in reactive mode. That includes getting your TMCH validation so you can participate in Sunrise periods for new domain launches and considering applying for your own .brand TLD.
GlobalBlock by the numbers
Instead of waiting for abuse to occur and reacting after the fact, GlobalBlock prevents registrations from happening in the first place. Since launching in March 2024, the BSA has partnered with registry operators worldwide to create the world’s first unified domain blocking service that covers 62% of the world’s top-level domains, including 75% of the world’s generic TLDs (gTLDs).
840+
Top-level domains covered
10M+
Domains protected
10,000+
Domains recovered via Priority AutoCatch
That last number is important. A significant number of those domains were registered before GlobalBlock existed. When third parties let them lapse, Priority AutoCatch automatically picked them up before they could be reclaimed by drop-catching services or re-enter the aftermarket. This feature is a huge win for brand protection teams.
How GlobalBlock provides prevention at scale
Microsoft describes the shift this way in the report: “Historically, our domain portfolio included a collection of defensive registrations accumulated over decades, with a tremendous annual carrying cost. GlobalBlock+ has enabled us to begin reducing our defensive domain portfolio while achieving a level of domain name protection previously unattainable.”
A global luxury fashion brand put it in enforcement terms: “Since purchasing GlobalBlock, we’ve reduced the number of abusive reports from our monitoring services and benefited from significant reductions in our enforcement activities. By shifting from reactive litigation to proactive protection, the burden on our organization has been dramatically reduced.”
This is the direction the industry is moving. The brands that navigate AI domain abuse most effectively will be the ones that invest now in prevention infrastructure that raises the cost of targeting them in the first place.
Find out how much of your brand is currently exposed.
Our team can run a Domain Vulnerability Report showing which domains containing your brand name are already registered, which are available, and where your coverage has gaps.