Taylor v Killer Queen LLC [2026] HCA 5 (the Case) is the final judgment given in the trade mark dispute between Australian designer, Katie Taylor, and US pop star, Katy Perry. The High Court allowed the appeal and affirmed that reputation under s 60 of the Trade Marks Act 1995 (Cth) (the TM Act) must relate to particular goods or services, and that personal reputation does not automatically constitute trade mark reputation across all categories.

I FACTS

In 2007, Katie Taylor (nee Perry) (henceforth referred to as Ms Taylor), established her clothing label under the name “Katie Perry”. Ms Taylor registered this as the domain name in May 2007, and sold clothing under the domain name from this period. On 29 September 2008, Ms Taylor applied to register the word mark “Katie Perry” in Class 25 (clothing), which became the Designer Mark.

Simultaneously, from 2002 onwards, Katheryn Hudson had performed under the stage name Katy Perry (henceforth referred to as Ms Hudson). Ms Hudson had garnered substantial recognition in Australia by September 2008 through musical song releases.

In mid-2009, Ms Hudson’s representatives were made aware of Ms Taylor’s trade mark, and were concerned that the purpose of the trade mark was to obtain an advantage based on Ms Hudson’s reputation. Subsequently, Ms Hudson’s representatives served correspondence upon Ms Taylor stating that she was to rescind her trade mark application, cease using the name “Katie Perry” on clothing, and to abstain from using any similar name hereafter. Ms Hudson’s representatives offered settlement terms to coexist, which were refused by Ms Taylor.

On 26 June 2009, Ms Hudson sought an Australian trade mark for “Katy Perry” in Classes 9 and 41 (music-related goods and services). Ms Hudson also sought a trade mark under Class 25 (clothing), but this was subject to an objection based on Ms Taylor’s earlier registration. As such, Ms Hudson deleted Class 25 from her trade mark application.

However, she proceeded to sell Katy Perry-branded merchandise in Australia, including clothing, headgear and footwear. Ms Hudson and her representatives were aware of the trade mark infringement risk, but considered there was no risk of anyone in Australia being confused because of her earlier overseas adoption and stronger reputation in the mark.

In September 2019, Ms Taylor’s representatives served a letter of demand to Ms Hudson’s representatives. Following Ms Taylor’s subsequent lawsuit for infringement on 20 December 2019, Ms Hudson filed a cross-claim seeking to rectify the register and cancel the “Katie Perry” trade mark.

II ISSUES BEFORE THE COURT

The High Court provided three central issues for determination;
I. Whether, under s 88(2)(a) read with s 60, the use of Ms Taylor’s Mark at the priority date (29 September 2008) would be likely to deceive or cause confusion because of the reputation of Ms Hudson’s Mark,
II. Whether, under s 88(2)(c), the use of Ms Taylor’s Mark was likely to deceive or cause confusion because of the circumstances applying as at 20 December 2019 (the date the rectification application was filed), and
III. Whether, if either ground was established, the discretion in s 89 was enlivened.

III HIGH COURT’S JUDGEMENT

Gordon A-CJ and Beech-Jones J would have dismissed the appeal, agreeing with the Full Court that both grounds for rectification were established and that the discretion under s 89 was not enlivened. On the s 60, their Honours accepted that Ms Hudson’s Mark had acquired a reputation in Australia before the priority date in relation to music and entertainment, based on the significant chart success of Ms Hudson’s singles and album, extensive media coverage, and the inference that the album had achieved “gold” status by 15 October 2008. They held that the use of Ms Taylor’s Mark would be likely to deceive or cause confusion because of the similarity of the marks and the “commonly accepted practice of pop stars selling merchandise including clothing”. Their Honours emphasised that evidence of actual confusion is not required; the question is whether there is a “real, tangible danger” of deception or confusion.

On s 88(2)(c), Gordon A-CJ and Beech-Jones J held that the inquiry is necessarily directed to the notional normal and fair use of the mark, not its actual use. The statutory monopoly conferred by registration extends to all goods within the specification, and the registered owner's failure to exploit the monopoly fully cannot limit the scope of the inquiry. By 20 December 2019, Ms Hudson had achieved international stardom and her mark had a significant reputation, including in relation to clothing merchandise, such that the likelihood of deception or confusion was even greater than at the priority date. The absence of evidence of actual confusion did not detract from this conclusion, given Ms Taylor’s very modest business presence. On the s 89 discretion, their Honours held that the word “act” does not require blameworthy conduct; something more than the mere act of registration is required, but the inquiry is whether the registered owner's conduct caused or substantially contributed to the ground arising. As at the priority date, Ms Taylor’s act of applying for registration with knowledge of Ms Hudson, her reputation, and the common practice of pop stars selling merchandise substantially contributed to the ground under s 88(2)(a). As at 20 December 2019, additional acts or faults included Ms Taylor’s encouragement of the association between herself and Ms Hudson through social media posts referencing the pop star, and her failure to enforce her statutory monopoly for over ten years. The discretion was therefore not enlivened.

Steward J agreed generally with Jagot J that the grounds for rectification were not established, but provided an additional and independent reason for rejecting the s 88(2)(c) ground. Even assuming a likelihood of deception or confusion existed by 2019, it would have been the product of Ms Hudson’s own persistent infringement of Ms Taylor’s registered mark. Steward J characterised Ms Hudson’s conduct as that of an “assiduous infringer”. The respondents had sold "Katy Perry" branded clothing in Australia from 2009 onwards, with full knowledge of Ms Taylor’s registered mark and in circumstances where Ms Hudson had voluntarily excluded class 25 goods from her own trade mark registration. Steward J held that it would be an “extraordinary result” for s 88(2)(c) to permit an applicant for rectification to rely on a likelihood of confusion that was itself the product of their own wrongful conduct. The phrase “likely to deceive or cause confusion” in s 88(2)(c) must be construed in a restricted way so as not to include deception or confusion caused by the infringing conduct of the applicant for rectification. Section 89 does not address this situation, as it is directed to the conduct of the registered proprietor rather than that of an assiduous infringer.

Gleeson J agreed that the appeal should be allowed and that orders should be made in the terms proposed by Jagot J. The Full Court had wrongly conflated the reputation of the “Katy Perry” mark as a trade mark with the personal reputation of Katy Perry as an individual, and had incorrectly treated the mark as having a “reputation in clothes” despite the absence of any use in respect of clothing before the priority date. Gleeson J held that the reputation of Ms Hudson’s Mark before the priority date was in respect of pop music concerts and pop music, not clothing. Applying the notional normal and fair use test, an ordinary customer who was aware of Ms Hudson’s Mark would expect clothes connected to that mark to convey an association with Katy Perry as a pop star and would dismiss the possibility of a shared trade source where clothes were not sold in a context identifying such an association. The likelihood of confusion was therefore remote. On s 88(2)(c), Gleeson J agreed that the test applies to notional use, but found that the increased recognition of Katy Perry and Ms Hudson’s Mark by 2019 did not establish the likelihood of confusion for the same reasons as under s 60. The courts below had made no finding that the volume of sales of clothing under Ms Hudson’s Mark demonstrated that it had acquired a reputation in connection with clothes. Since neither ground for rectification was established, the s 89 discretion did not arise.

Jagot J held that a trade mark can acquire a reputation in Australia only in respect of particular goods or services, and that such a reputation does not automatically extend to other goods or services, though use of a mark may still be likely to deceive or cause confusion because of a reputation in different goods or services as a question of fact. She further held that the s 88(2)(c) inquiry extends to notional normal and fair use rather than actual use, that s 89 is the sole source of power not to rectify the Register, and that the mere act of applying for registration cannot constitute an “act or fault” under s 89(1), though the circumstances of the application may.

Jagot J found that before the priority date Ms Hudson’s Mark had acquired a reputation in Australia in respect of entertainment and recorded music, but not clothing, and that no “Katy Perry” branded clothing had been sold in Australia before that date. The primary judge’s conclusion that the respondents had not discharged their onus of proving a real, tangible danger of confusion was not unreasonable or untenable. An ordinary person aware of Ms Hudson’s Mark would expect merchandise to explicitly highlight its connection to the artist; the bare use of “Katie Perry” on clothing without any such signifier would not cause confusion. The Full Court had attributed errors to the primary judge that were not apparent from their reasons, and had impermissibly substituted its own opinion for that of the trial judge. On s 88(2)(c), the complete absence of evidence of actual confusion over ten years was sufficient to support an inference that confusion was not likely, and the trade channels used by Ms Taylor meant the marks would overwhelmingly be seen rather than heard, reducing the significance of their aural similarity. Since no ground for rectification was established, no question arose under s 89.

IV FINAL ORDERS

The High Court ordered that the appeal was to be allowed with costs. Further, that orders 1, 3, 4, 5, 6, 7 and 8 of the Full Court were to be set aside, and that the matter was to be remitted to the Full Court in relation to grounds 5, 6 and 7 and any other outstanding issues, including costs.

V SIGNIFICANCE AND IMPLICATIONS

The Case is a significant decision for Australian trade mark law for several reasons;

1. The majority confirmed that a trade mark can acquire a reputation in Australia only in respect of particular goods or services, and that a reputation in one class of goods or services does not automatically extend to other classes. This is an important constraint on the scope of s 60 of the TM Act and prevents the reputation of a famous mark in one field from automatically blocking registration in unrelated fields.

2. Further, the Court unanimously confirmed that the likelihood of deception or confusion under both provisions is assessed by reference to the notional normal and fair use of the registered mark across the full range of goods in the specification, not merely the registered owner's actual use. This prevents registered owners from resisting rectification by pointing to a narrow actual use of a broadly specified mark.

3. The majority’s approach gives significant weight to a prolonged absence of evidence of actual confusion where both marks have been in use in the market. Jagot J held that ten years of co-existence without any evidence of confusion, was sufficient to support an inference that confusion was not likely.

4. Steward J’s reasons raise the important question of whether s 88(2)(c) should be construed to exclude circumstances caused by the wrongful conduct of the applicant for rectification. While Jagot J declined to resolve this question, Steward J held that Parliament could not have intended to reward an applicant who itself had created the confusion through persistent infringement. This reasoning may prove influential in future cases where an applicant for rectification itself has used a mark in breach of the registered owner's rights.

5. The Court confirmed that the mere act of applying for registration of a trade mark cannot, without more, constitute an “act or fault” for the purposes of s 89(1). However, the circumstances in which the application was made, including the applicant's state of mind and knowledge, may constitute such an act.

VI FOOTNOTES

Trade Marks Act 1995 (Cth) s 60.

Taylor v Killer Queen LLC [2026] HCA 5 , 60-62.

Ibid, 7, 63–64, 72.

Ibid, 54, 67.

Ibid, 82–86.

Ibid, 83.

Ibid, 89–94.

Ibid, 91–92.

Ibid, 97–100.

Ibid, 103–104.

Ibid, 108–111.

Ibid, 117, 163.

Ibid, 127.

Ibid, 128.

Ibid, 156–157.

Ibid, 157.

Ibid, 159.

Ibid, 211.

Ibid, 189.

Ibid, 197.

Ibid, 200.

Ibid, 209.

Ibid, 215.

Ibid, 216.

Ibid, 217.

Ibid, 218.

Ibid, 220.

Ibid, 255, 263, 282.

Ibid, 283.

Ibid, 277, 284–285.

Ibid, 281.

Ibid, 293–295.

Ibid, 296.

Ibid, 301.

Ibid.

Ibid.

Ibid, 215–216, 238.

Ibid, 55, 85, 218, 243.

Ibid, 294.

Ibid, 156–157.

Ibid, 220, 247–248.

Ibid, 99, 248.