'Never Really Over': the Katy Perry v Katie Perry battle continues

9 Minute read 18.03.2026 John Fairbairn, Jaimie Wolbers, Georgia Murray

The High Court, by a 3-2 majority, partially determines this long-running dispute, refusing to cancel the designer's trade mark registration.

Key takeouts

The dispute is ongoing, with the matter being remitted to the Full Federal Court to consider whether the remedies awarded by the primary judge were appropriate, and to determine the issue of costs.

The litigation history of this matter is complex, but it demonstrates the importance of enforcing rights against potential trade rivals early - delay can lead to substantially increased complexity and costs.

Businesses must conduct due diligence, including considering a third party's international and local reputation, before investing in a brand.

The latest chapter in the long-running trade mark saga featuring Australian fashion designer, Katie Taylor (Ms Taylor), and well known American pop star, Katheryn Hudson (ie 'Katy Perry') (Ms Hudson), has now been written, with the High Court of Australia delivering their judgment in Taylor v Killer Queen LLC [2026] HCA 5.

The dispute between Ms Taylor and Ms Hudson was described by the Full Court of the Federal Court of Australia (the Full Court) in Killer Queen, LLC v Taylor [2024] FCAFC 149 as 'an unfortunate one' between 'two enterprising women in different countries'.

Practical implications of the High Court's decision

The decision, particularly the reasoning of Jagot J, highlights several practical points for trade mark owners and applicants, including that:

First, a trade mark can only acquire a reputation in Australia in respect of particular goods or services, a general reputation or celebrity status alone is insufficient.
Second, the concept of 'notional normal and fair use' has important limits: while it encompasses all legitimate uses within the scope of a registration, it does not extend to uses that would amount to trading off another party's reputation.
Third, the absence of evidence of actual confusion (while not determinative), can be a significant factor, particularly where both marks have co-existed in the market for an extended period and the party seeking cancellation bears the burden of proof. Fourth, the strength of a well-known mark may in some cases reduce rather than increase the likelihood of confusion, as consumers familiar with the well-known mark may recognise that an unfamiliar use is unconnected to it.
Finally, the decision underscores the importance of appellate courts exercising restraint in trade mark disputes involving evaluative findings of fact, where the primary judge has had the benefit of hearing the entirety of the evidence.

By way of background:

Ms Taylor held a registered trade mark in Australia for the words "Katie Perry" in connection with the following goods in class 25: Clothes. The priority date of the trade mark was 29 September 2008.
At first instance (Taylor v Killer Queen, LLC (No 5) [2023] FCA 364), the primary judge found that Ms Hudson's related corporate entities had infringed Ms Taylor's registered trade mark in connection with sales of concert and album related merchandise and dismissed Ms Hudson's application to cancel that trade mark. Ms Hudson herself was found not to have infringed the trade mark by reason of the "own name" defence. The primary judge made a number of declarations and ordered injunctive relief, with the parties to negotiate orders in connection with a hearing on the issue of the quantification of damages.
Ms Hudson's corporate entities appealed that decision, and the Full Court overturned the primary judge's decision in connection with the cross-claim, ordering that Ms Taylor's KATIE PERRY trade mark registration be cancelled. This had flow on effects to the issue of relief, with the relief that had been ordered being set aside.
Ms Taylor then made an application for special leave, and was permitted to appeal the Full Court's decision in the High Court.
On 11 March 2026, the High Court of Australia delivered its much anticipated judgment, allowing Ms Taylor's appeal by a 3-2 majority and setting aside the vast majority of the Full Court's orders. The matter has now been remitted to the Full Court for determination of outstanding matters including the appropriate relief to be granted in connection with infringement (including to consider whether delay (or laches) is a disentitling factor in Ms Taylor obtaining injunctive relief or an account of profits) and costs.

What was the High Court required to determine?

The dispute between Ms Taylor (the designer) and Ms Hudson (the pop star) broadly involves two main components:

Ms Taylor's claim that the respondents had infringed her KATIE PERRY trade mark by selling KATY PERRY branded clothing in Australia under section 120 of the Trade Marks Act 1995 (Cth) (the Act); and
Ms Hudson's (and her related corporate entities') cross-claim seeking cancellation of Ms Taylor's trade mark under section 88 of the Act.

The cross-claim for cancellation raised three key issues:

(a) whether Ms Taylor's registration could have been opposed under section 60 of the Act on the basis that Ms Hudson's KATY PERRY mark had acquired a reputation in Australia before the priority date and, due to that reputation, the use of the KATIE PERRY trade mark by the designer in relation to clothes would be likely to deceive or cause confusion (section 88(2)(a) of the Act);

(b) whether, due to the circumstances applying at the time the application for rectification was filed (20 December 2019), the use of the KATIE PERRY trade mark by the designer was likely to deceive or cause confusion (section 88(2)(c) of the Act); and

The High Court's appeal concerned only the cancellation cross-claim - that is, the issues arising under sections 60, 88(2)(a), 88(2)(c) and 89 of the Act. The Full Court's findings on infringement, including the scope of Ms Taylor's registration (ie whether 'clothing' included footwear and headgear), joint tortfeasorship, and the availability of defences including the 'own name' defence under section 122(1)(a)(i) and the 'honest concurrent use' defence under sections 44(3) and 122(1)(fa) of the TM Act, were not in issue before the High Court.

The High Court's decision

As noted above, the High Court allowed Ms Taylor's appeal by a 3-2 majority and set aside the Full Court's orders cancelling Ms Taylor's trade mark. While all five judges agreed on several questions of statutory construction (including that the relevant 'use' under sections 60 and 88(2)(c) is notional normal and fair use, and that section 89 is the sole source of the Court's discretion not to rectify the Register if the grounds in section 88 are otherwise established) they differed in the application of the law to the facts.

Reputation and likelihood of confusion at the priority date (sections 60 and 88(2)(a))

All five judges agreed that the pop star's KATY PERRY mark had acquired a reputation in Australia before the priority date (29 September 2008) in respect of entertainment and music. However, the judges differed on whether that reputation extended to clothing.

The majority (Gleeson, Steward and Jagot JJ, in separate judgments) held that the Full Court erred in finding that the 'common practice' of pop stars selling merchandise including clothing could extend the reputation of the KATY PERRY mark to clothing. Jagot J emphasised that not a single item of KATY PERRY branded clothing had been sold in Australia before the priority date (being 29 September 2008), and a common practice of pop stars selling merchandise could not give the mark a reputation in goods it had never been used on in this jurisdiction. The dissent (Gordon A-CJ and Beech-Jones J) accepted that the reputation did not extend to clothing as such, but held this did not matter because the reputation in music and entertainment was sufficient to ground the section 60 inquiry.

On the question of whether the use of Ms Taylor's registration would be likely to deceive or cause confusion due to the KATY PERRY mark's reputation, the majority held that it would not. Gleeson J reasoned that ordinary customers aware of the KATY PERRY mark would expect clothes sourced from Katy Perry to bear some association with her as a pop star, such as images, song names or tour details, beyond the standalone name 'Katie Perry'. In the absence of such features, consumers would dismiss the possibility of a shared trade source. Jagot J similarly held that an ordinary person aware of the KATY PERRY mark's reputation in entertainment would expect any clothing connected to Katy Perry to highlight that connection, and clothing bearing only 'Katie Perry' with no such connection would not create a real, tangible danger of confusion.

Importantly, Jagot J held that the concept of 'notional normal and fair use', which considers all the ways a trade mark could legitimately be used within the scope of its registration, does not extend to illegitimate or improper uses by the trade mark owner. Given that the placing of Katy Perry's image on clothing or selling clothing at a Katy Perry concert would not be a 'normal and fair' use by the designer, such potential uses could not be considered for the purposes of section 88(2). Jagot J also held that the primary judge did not err in treating the absence of any evidence of confusion over many years as a 'powerful factor', noting that the burden of proving the likelihood of confusion rested on the party seeking cancellation.

Likelihood of confusion at the date of the rectification application (section 88(2)(c))

The majority also held that as at 20 December 2019, there was no likelihood of confusion arising in connection with the designer's continued use of her registered KATIE PERRY trade mark:

Gleeson J held that the evidence did not support a conclusion that the increased recognition of Katy Perry led to a likelihood of confusion that would not have resulted from the strength of the reputation at the priority date (29 September 2008).
Jagot J held that ten years of co-existence with no evidence of confusion, despite the Katy Perry parties' attempts to find such evidence through discovery, supported an inference that confusion was not likely. The strength of the KATY PERRY mark by 2019 was capable of supporting the conclusion that consumers would, after a moment's reflection, recognise the marks were unconnected.
Steward J agreed with Jagot J's reasoning on this issue. Jagot J further emphasised that the Full Court should not have interfered with the primary judge's evaluative findings, given the evaluative nature of the statutory exercise and the primary judge's advantage of hearing the entirety of the evidence.

The dissenting judgment found that by 20 December 2019, the likelihood of confusion was even greater than at the priority date given Ms Hudson's increased international fame, the similarity of the marks, and the common practice of pop stars selling merchandise. The dissent considered that the absence of actual confusion did not detract from this conclusion given the limited use of Ms Taylor's trade mark.

The 'assiduous infringer' principle

Steward J, while agreeing with Jagot J that the grounds for cancellation were not made out on the evidence, advanced an additional reason for dismissing the section 88(2)(c) ground. His Honour held that section 88(2)(c) of the Act must be construed so as not to include deception or confusion caused by the infringing conduct of the applicant for rectification. Steward J characterised Ms Hudson and her corporate entities' conduct as that of persistent, assiduous infringers who sold KATY PERRY branded clothing in Australia for years with full knowledge of Ms Taylor's registered mark, and held that Parliament could not have intended to reward such conduct by permitting them to rely on the consequences of their own infringement to cancel Ms Taylor's trade mark. Jagot J preferred to express no view on this question.

Section 89 - Discretion not to cancel

Section 89 gives the court the discretion to decide not to rectify the register where the likelihood of deception of confusion has not been caused by any act or fault of the registered owner. All five judges agreed that section 89 is the sole source of power not to rectify the Register where a ground in section 88 is otherwise established.

As the majority found no ground for cancellation was established, the section 89 discretion did not arise for their consideration in this case. However, Jagot J addressed the construction of section 89(1) of the Act, holding that the mere fact of having applied for and obtained registration of a trade mark cannot itself be an 'act or fault' within the terms of section 89(1).

What happens next?

The High Court remitted the matter to the Full Court for determination of three outstanding grounds of appeal (grounds 5, 6 and 7) which the Full Court had not addressed in its original decision, having found that Ms Taylor's trade mark should be cancelled.

Now that the High Court has restored Ms Taylor's trade mark, the Full Court will need to determine:

(a) ground 5 - whether the primary judge erred in rejecting the respondents' laches defence (the respondents argued that Ms Taylor's delay in commencing proceedings disentitled her to relief);

(b) ground 6 - whether the primary judge erred in awarding additional damages against Kitty Purry under section 126(2) of the Act on the basis that its infringement was flagrant; and

The Full Court will also need to determine the costs of the appeal and of the proceeding before the primary judge. Accordingly, while the High Court's decision resolves the cancellation dispute in Ms Taylor's favour, the broader proceedings between the parties are not yet concluded.

The High Court's decision reinforces the importance of the statutory monopoly conferred by trade mark registration and the significant evidentiary burden on a party seeking to cancel a registered trade mark.

When relying on a competing reputation as a basis for cancellation, it is critical that there be evidence of that reputation in relation to particular goods and services in Australia. Evidence of a more general reputation in respect of other goods and services will likely be insufficient.

Businesses must conduct due diligence, including considering third party reputations in Australia and other jurisdictions, before investing in a brand or commencing trade in Australia.

The dispute also highlights the value of maintaining and promptly enforcing registered trade marks, as the delay in Ms Taylor commencing proceedings to enforcing her registered trade mark has significantly contributed to the cost and complexity of this matter.