Request a three-member panel to defend a .ai domain: what panels actu…

Request a three-member panel to defend a .ai domain: what panels actu. UDRP and ccTLD domain recovery and defense across .ai. Email the firm to assess your cas…

A complaint lands. The zone is .ai — Anguilla's country-code extension, now synonymous globally with artificial intelligence and attracting a wave of trademark claims from technology companies that never registered a .ai domain until a stranger held the one they wanted. You registered the domain legitimately. The complainant is well-resourced. The single panelist assigned to your case could go either way. So you consider the one procedural lever available to the respondent: requesting a three-member panel.

Under the UDRP — which applies to .ai domains administered through WIPO — a respondent may request a three-member panel in place of the default single panelist by paying the difference in the filing fee, bringing the total forum cost to USD 4,000 for a single-domain case at WIPO. That procedural choice can reshape the outcome. Three-member panels are statistically more deliberate, more likely to produce dissents, and more likely to issue reverse domain name hijacking (RDNH) findings when the complaint is weak. The decision to request one — and how to build the record around it — is the subject of this analysis.

This page examines the governing rules for .ai disputes, the safe harbors and legitimate-interest record that decide the merits, what evidence panels weight most heavily, and when a three-member request is the right call versus a distraction.

How does the UDRP apply to .ai domains, and what does that mean for respondents?

The .ai ccTLD is administered by the government of Anguilla, which has appointed WIPO as its dispute-resolution provider, making .ai one of the more than 87 ccTLDs that operate under WIPO's dispute procedures. In practical terms, a complainant filing against a .ai domain follows the same three-element UDRP test it would use for a .com — with the same evidentiary standards, the same panel pool, and the same remedies (transfer or cancellation; no damages). This convergence is significant for respondents: the decades of UDRP precedent on legitimate interest, bad faith, and RDNH all carry weight in .ai proceedings.

The three elements a complainant must prove, under Paragraph 4(a) of the UDRP, are: (1) the domain is identical or confusingly similar to a trademark in which the complainant has rights; (2) the respondent has no rights or legitimate interests in the domain; and (3) the domain was registered and is being used in bad faith. All three must be established. A complainant who can satisfy element one — often easy if it holds a registered mark — still faces elements two and three, and the cumulative nature of the "registered and used" bad-faith test is where many .ai complaints against genuine domain investors or developers falter.

Why does the .ai zone attract a disproportionate volume of aggressive complaints right now? The AI industry boom has driven technology companies to seek .ai registrations retroactively, sometimes years after a domain investor acquired the name for its obvious commercial value in an emerging sector. Panels have consistently recognized that registering a short, generic, or acronymic domain name does not establish bad faith merely because a trademark holder later emerges with a claim. That principle is tested repeatedly in .ai proceedings, and it is one reason why the three-member panel option matters — closer scrutiny of a marginal complaint benefits a respondent with a coherent registration story.

In our practice, we regularly advise respondents who receive a UDRP complaint against a .ai domain they acquired in good faith. For a preliminary read on whether the complaint meets the three-element threshold, contact info@cognomenlaw.com.

What are the Paragraph 4(c) safe harbors, and how do they apply in .ai disputes?

Paragraph 4(c) of the UDRP provides three recognized safe harbors that, if demonstrated, establish a respondent's rights or legitimate interests and defeat the complaint at element two. The three are: (a) bona fide use of the domain in connection with a genuine offering of goods or services before any notice of the dispute; (b) the respondent being commonly known by the domain name; and (c) legitimate noncommercial or fair use, without intent for commercial gain to mislead consumers or tarnish the mark.

In .ai disputes, safe harbor (a) is the most frequently invoked and the most fact-intensive. Panels look at whether the use was genuine and predated notice — not whether it was commercially successful. A respondent who parked the domain on a pay-per-click (PPC) page faces a harder argument, particularly if those ads target the complainant's industry. But a respondent who can show development activity, a business plan with contemporaneous documentation, or a holding consistent with a coherent investment thesis in AI-related names stands on firmer ground. The distinction panels draw is between speculative registration (legitimate) and registration targeting a specific mark owner's goodwill (abusive).

Safe harbor (b) — being commonly known by the name — rarely applies to domain investors but becomes relevant where a registrant has operated under a business name or brand that the disputed domain reflects. Safe harbor (c) is the domain of commentary, criticism, and fan sites, and has limited application in the .ai context where most contested names have obvious commercial value.

Building the legitimate-interest record before filing a response is the single most consequential task in defending a .ai domain. That means assembling: registration date and the registrant's knowledge (or lack of knowledge) of the complainant's mark at that date; evidence of any development, monetization, or holding consistent with an investment strategy; correspondence predating the complaint; WHOIS history; and any communications in which the registrant responded to inquiries about the domain without making an extortionate demand. We have defended .ai registrations successfully by reconstructing precisely this kind of contemporaneous record, even when the domain had been held passively.

When is requesting a three-member panel the right strategic choice?

A respondent may, within its response, request a three-member panel instead of the single panelist the complainant's filing fee covered. If the respondent makes that request, the parties generally split the higher three-member fee — meaning the complainant pays its share (already paid) and the respondent pays the difference bringing the total WIPO forum cost to USD 4,000 for a single-domain case. The respondent's share of that uplift is typically half the difference between the single-panel fee (USD 1,500) and the three-member fee (USD 4,000): approximately USD 1,250 in addition to the respondent's own legal costs.

Is that cost justified? It depends on three variables. First: how strong is the complaint? A complaint that is clearly abusive — filed by a party without a credible trademark, or relying on a mark acquired after the domain's registration date, or making arguments panels have consistently rejected in analogous .ai cases — is a candidate for an RDNH finding. RDNH findings are more common from three-member panels, because a dissent from one of three panelists signals genuine disagreement, and the deliberative process of three panelists produces more careful reasoning about whether the complainant's filing was brought in bad faith. Second: how high are the stakes? A .ai domain with meaningful commercial value — whether as an investment asset or as active infrastructure for a business — justifies the additional forum cost. Third: what is the respondent's litigation posture? A three-member panel is not a guarantee of a better outcome; a weak response will lose before three panelists as readily as before one.

The consensus view in the UDRP community is that three-member panels are more likely to scrutinize complainant conduct carefully, more willing to engage with nuanced legitimate-interest arguments, and more likely to find RDNH where the evidence supports it. The contrary view — held by some practitioners — is that in a straightforward case the single panelist is as likely to reach the correct result, and that requesting a three-member panel signals to the panel that the respondent expects a fight, potentially shading the panel's first impressions. In our view, that concern is overweighted: a well-reasoned response speaks for itself regardless of the panel's composition, and where an RDNH finding is a realistic objective, the three-member route is nearly always preferable.

What evidence decides the outcome in a defended .ai domain dispute?

The evidentiary record in a .ai domain defense is built around two questions that panels ask in sequence: what did the respondent know at the time of registration, and what has the respondent done with the domain since? Bad faith under the UDRP is not judged at the time of the complaint — it must have been present at registration. This is a critical protection for respondents who registered .ai domains before the complainant's mark became prominent, or before the complainant itself existed in its current form.

Panels weigh the following evidence categories most heavily. Registration date versus mark priority: if the complainant's trademark registration postdates the domain's registration, the respondent has a strong chronological argument unless the complainant can show common-law rights predating the domain. Market context at registration: for .ai domains, the question of whether "AI" was in the public domain as an acronym for artificial intelligence at the time of registration — or whether the domain's short form was being registered for other reasons — shapes the panel's view of intent. Monetization conduct: PPC pages targeting the complainant's industry are damaging; generic advertising or a blank holding page is far less so. Offer-to-sell evidence: if the registrant solicited the complainant with a demand, that is significant bad-faith evidence; if the complainant approached the registrant and the registrant named a price, panels treat that differently, though it is not entirely safe harbor.

In a recent matter (a .ai investment domain dispute, spring 2025), we assembled a registration record showing our client had acquired the domain as part of a portfolio of AI-sector names before the complainant's mark achieved significant market penetration. The panel found no bad faith at registration and denied the complaint. The three-member panel we requested produced a unanimous decision — and a clear statement on what the complainant had failed to prove — that the registrant has since used to manage inbound inquiries about the name's status.

Two additional evidence categories warrant attention. First, the respondent's response to pre-complaint solicitations: if the complainant's representatives approached the registrant before filing and the registrant's reply was measured — neither extortionate nor evasive — panels treat that correspondence as a legitimacy signal. Second, portfolio character: a respondent who holds dozens or hundreds of AI-sector domain names, registered consistently across time, is more credible as an investor than one who holds a single domain that happens to match the complainant's mark.

How do panels assess RDNH, and when is it a realistic objective in a .ai defense?

Reverse Domain Name Hijacking is a finding that the complaint was brought in bad faith — that is, that the complainant used the UDRP to attempt to deprive a legitimate registrant of a domain it had no proper claim to. An RDNH finding carries no monetary penalty; its value is reputational and, for the respondent, confirmatory. But it also creates a public record that deters future abusive filings and signals to the domain industry that the complainant overreached.

Panels have consistently held that RDNH is available where the complainant knew or should have known it could not succeed — for example, where the domain predates the mark by years, where the complainant has no evidence of targeting or bad faith but files anyway, or where the complaint contains misrepresentations about the strength or scope of the trademark. The threshold for an RDNH finding has been described by panels as requiring a showing of something beyond a lost case: the complainant must have filed with an improper purpose or with willful blindness to the obvious legitimacy of the registrant's interest.

In the .ai context, RDNH is a realistic objective in several recurring scenarios. A technology company that filed a trademark application after the domain's registration date and then filed a UDRP complaint within months of obtaining the registration — citing that freshly obtained mark — is a credible RDNH candidate. A company that sent a pre-complaint letter offering to purchase the domain, received a reasonable reply, and then filed a UDRP complaint presenting that exchange as evidence of bad faith is another. A company that filed against a domain holding generic or descriptive terms on the basis of a highly stylized trademark is a third.

We regularly advise registrants in these situations. Pursuing RDNH requires documenting not only the respondent's legitimate interest but also the complainant's filing conduct — what the complainant knew before filing, what the pre-complaint correspondence shows, and whether the trademark arguments were objectively tenable. Three-member panels are the better venue for this argument: a single panelist may decline to issue an RDNH finding where the complaint was plausible if weak, whereas three panelists can reach a majority finding even if one dissents.

If the complaint you have received contains arguments that are demonstrably foreclosed by the registration timeline or by prior panel decisions, an RDNH finding may be within reach. Email info@cognomenlaw.com to assess whether the complaint meets that threshold.

How does defending a .ai domain compare to defending a .com or a .uk domain?

The right route depends on the zone, and the .ai zone carries specific features that shape a defense strategy. Under the UDRP — which governs both .com and .ai disputes at WIPO — the procedural posture and legal test are identical. A respondent defending a .ai domain uses the same response format, the same 4(c) safe harbors, the same bad-faith chronology, and the same RDNH framework as a .com respondent. The key difference is factual: the AI industry context means that many .ai complaints are filed by technology companies who registered or obtained marks after the domain was acquired, a timing issue that frequently determines the outcome.

Contrast the .uk zone. A .uk domain dispute proceeds under Nominet's DRS — a distinct procedure with a mandatory mediation stage before any expert decision. The DRS test is "abusive registration," and crucially it reads "registered or used" abusively, a lower bar than the UDRP's cumulative "registered and used in bad faith." A respondent defending a .uk domain who has used the name commercially since registration faces a different argument from a .ai respondent, because Nominet's test can be satisfied if current use is abusive even if registration was innocent.

The .de zone presents yet another picture. There is no UDRP for .de. Disputes go to the German courts, and DENIC offers a dispute entry that blocks transfer while litigation proceeds — a protective step but not itself a decision on the merits. A respondent holding both a .com and a .ai in a contested name may face parallel proceedings in different systems simultaneously, each with different timelines and different evidentiary standards.

Where both a .ai and a .com are at issue, the UDRP allows a single complaint to cover multiple domains if the registrant is the same holder. A respondent in that position can request a three-member panel across both domains for a single elevated forum fee — a meaningful efficiency where the defense arguments are common across the zones.

In another matter we handled (a dual .com/.ai challenge, autumn 2024), the complainant filed a single complaint covering both domains. We requested a three-member panel, presented a unified legitimate-interest record, and secured a denial on both domains. The panel's written decision addressed the registration timeline across both zones and found no credible evidence of targeting the complainant's mark.

What is the realistic next step for a respondent in a .ai domain dispute?

A respondent who receives a UDRP complaint against a .ai domain has 20 days from the date of commencement to file a response. Missing that deadline does not automatically mean the domain is transferred — a panel can still deny a complaint on the merits even without a response — but defaulting removes the respondent's opportunity to present the legitimate-interest record and eliminates any realistic path to an RDNH finding. Acting within the response window is the foundational step.

The realistic decision tree looks like this. If the complaint is clearly abusive — the mark postdates the domain, the arguments are objectively weak, and the pre-complaint conduct is problematic — request a three-member panel and build the response around both legitimate interest and RDNH. If the complaint is borderline — the complainant has a plausible mark but the registration was innocent — decide the panel composition based on the commercial value of the domain and the respondent's tolerance for the additional forum cost. If the complaint is strong — the respondent's conduct is hard to distinguish from targeting — a three-member panel does not change the calculus materially, and the response should focus narrowly on any safe harbor that is genuinely available.

What the three-member panel option is not: it is not a delay mechanism, it is not a guaranteed path to a better outcome, and it is not a substitute for a well-built evidentiary record. Panels are experienced practitioners who see the full range of respondent conduct. A request for a three-member panel accompanied by a response that fails to substantiate legitimate interest will lose, sometimes with a sharper analysis than a single panelist might have produced.

The AUDIENCE_MYTH worth addressing directly: the belief that a domain registrant cannot win against a large brand owner because the UDRP "favors complainants." The UDRP is a balanced procedure. Panels deny complaints routinely where the registrant presents a coherent legitimate-interest record and the complainant cannot establish the cumulative bad-faith test. In .ai disputes specifically, the registration-before-mark chronology is a genuine defense that panels respect. The three-member panel route maximizes the respondent's ability to present and have that record thoroughly considered.

Related at COGNOMEN

• UDRP respondent defense and RDNH — protecting your domain against abusive complaints across gTLD and ccTLD zones
• Defending an investment domain under US anticybersquatting rules — strategy and forum selection for portfolio registrants
• Case study — coordinated UDRP recovery across a domain group, strategy and outcome