The Apple AirTag MDL Goes to Argument on 24 September — and Apple Is Playing for a Denial

The Judicial Panel on Multidistrict Litigation hears oral argument on 24 September 2026 in Chicago on whether to centralize the federal Apple AirTag stalking actions as MDL No. 3199, and the unusual feature of the fight is that the defendant is not arguing about venue first — it is arguing that no Apple AirTag MDL should exist at all. Apple told the Panel in its 25 August 2026 opposition that it had already produced more than 285,000 pages of documents, sat for 39 plaintiff depositions, and settled 20 actions in the preceding two months, so that centralization would slow down cases that are closer to resolution than to discovery.

That is the whole case in one sentence, and it is a better argument than it sounds. Centralization under 28 U.S.C. § 1407 is not automatic: the Panel centralizes only where actions share common questions of fact and where transfer will serve the convenience of parties and witnesses and promote the just and efficient conduct of the litigation. The second half of that test is where a defendant with a nearly finished common discovery record and a working settlement channel does its damage.

What the JPML decides on 24 September 2026

The Panel is deciding one thing: whether the federal AirTag actions go to a single transferee judge for coordinated pretrial proceedings, or stay where they were filed. It is not deciding whether Apple is liable, whether the claims are good, or what any case is worth. A transfer order under § 1407 moves cases for pretrial purposes only; absent a waiver, each case returns to its originating district for trial.

Three positions are before the Panel:

Position	Who advances it	Requested outcome	Core argument
Centralize in Arizona	Moving claimants (4 August 2026 motion)	Transfer to the District of Arizona before Judge Diane Humetewa	Every action turns on the same AirTag design, the same corporate knowledge of misuse, and the adequacy of the same anti-tracking safeguards; duplicative discovery and inconsistent rulings are otherwise certain
No centralization	Apple (25 August 2026 opposition)	Deny the motion outright	Common discovery is largely complete; what remains is claimant-specific; informal coordination and active settlement are working, and an MDL would slow the more advanced cases
Centralize in Florida	Apple, in the alternative	Transfer to the Middle District of Florida	If the Panel centralizes, the transferee court should be one already hosting several of the pending actions

Will the JPML consolidate the Apple AirTag stalking lawsuits?

No public order answers that yet, and the Panel gave no signal in setting the matter for argument. The structural odds favor a grant — the Panel centralizes a clear majority of the products liability petitions it hears, and an inventory spread over 17 districts against one defendant on one product is the paradigm case for it — but Apple has built its opposition on the two grounds the Panel actually uses when it denies: that informal coordination among a small group of counsel is feasible, and that the common pretrial work is substantially complete. A ruling is expected within roughly one to two weeks of the 24 September argument.

The practitioner question is not which side has the better slogan. It is which of the Panel's two recurring denial rationales — "too few actions and informal coordination is feasible" and "the common discovery is already done" — has enough force against an inventory that claimants say numbers 27 actions across 17 district courts, and that is still growing.

What is the Apple AirTag MDL?

The Apple AirTag MDL is the proposed federal Multidistrict Litigation docketed at the Panel as MDL No. 3199, captioned In re: Apple AirTag Marketing, Sales Practices, and Products Liability Litigation. It does not yet exist. It is a request, filed on 4 August 2026, to consolidate the pending federal AirTag stalking actions before one judge for pretrial purposes under § 1407.

The underlying claims are a Mass Tort in the conventional products sense. Claimants allege that a person used an AirTag — a Bluetooth tracker Apple introduced in 2021 at a $29 price point — to follow them without consent, and that Apple is answerable in negligence and strict products liability for a Design Defect and a Failure to Warn: that the device's combination of size, price, and network-assisted precision made covert location tracking trivially easy, that Apple knew it, and that the safeguards it shipped deterred rather than prevented misuse. Reporting on the underlying complaints puts the volume of stalking-related reports Apple received at roughly 40,000 between April 2021 and April 2024.

Apple's answer on the merits is that the harm was inflicted by a third-party criminal actor, not by a defect, and that it built and then expanded the industry's anti-stalking tooling — culminating in the cross-platform Detecting Unwanted Location Trackers specification it developed with Google, proposed in May 2023 and shipped in May 2024 in iOS 17.5 and on Android 6.0 and later.

How a failed class action produced an MDL motion

The centralization motion is a direct consequence of a Rule 23 loss, and that sequence is the part the consumer coverage of this litigation consistently misses.

Hughes v. Apple, Inc., No. 3:22-cv-07668-VC, was filed in the Northern District of California in 2022 and assigned to Judge Vince Chhabria. On 15 March 2024 the court granted Apple's motion to dismiss in part: most theories were dismissed, but negligence and risk-benefit strict products liability claims survived for three of the named plaintiffs under California law. That ruling gave the litigation a merits spine.

Class Certification is where it stopped. In March 2026 Judge Chhabria declined to certify a class of alleged stalking victims, reported grounds being the variation among state laws and the individualized character of each stalking episode — the classic Predominance failure — together with a standing problem on classwide injunctive relief, since the named plaintiffs could not show they were likely to be tracked again. Reporting notes that the court compared the case at argument to the Uber sexual assault litigation, which is itself an MDL rather than a certified class.

The aggregation did not disappear when the class did; it changed shape. At least 16 individual actions, several filed pseudonymously, were docketed on 2 May 2026, and the count kept climbing through the summer. Plaintiffs' counsel from the Hughes case — Wade Kilpela Slade LLP — carried the inventory into individual filings. By early August, moving claimants told the Panel there were 27 related actions in 17 districts; contemporaneous reporting described at least 20 actions across at least 17 federal districts. The two counts are not reconcilable from public sources and are reported here as each source stated them.

Stage	Date	Event
Class complaint	2022	Hughes v. Apple, No. 3:22-cv-07668-VC, filed in N.D. Cal.
Pleadings	15 March 2024	Motion to dismiss granted in part; negligence and strict products liability claims survive for three named plaintiffs
Certification	March 2026	Class certification denied; state-law variation and individualized incidents defeat classwide treatment
Refiling wave	2 May 2026	At least 16 individual actions filed, several under pseudonym
Centralization sought	4 August 2026	§ 1407 motion filed; D. Ariz. before Judge Humetewa proposed; 27 actions in 17 districts identified
Opposition	25 August 2026	Apple opposes; cites 285,000+ pages produced, 39 plaintiff depositions, 20 settlements in two months; M.D. Fla. as alternative
Argument	24 September 2026	JPML hearing session, Chicago

Why does Apple oppose an AirTag MDL?

Apple opposes because centralization would cost it the two advantages it currently holds: a discovery record it has already paid for, and a settlement channel running at a pace it controls.

Its brief makes four moves. First, commonality is conceded in substance but drained of value — the common corporate discovery, Apple says, was done in Hughes, so a new MDL would mostly supervise claimant-specific work. Second, the remaining issues are individualized: who placed the tracker, what notifications the claimant's phone generated, what the claimant did, what injuries followed. Third, the number of parties and involved courts is small enough that informal coordination among a handful of firms is workable. Fourth — and this is the sharpest point — the parties settled 20 actions in two months, and centralizing "jeopardizes the parties' recent progress to global resolution" and risks slowing the more advanced cases.

Each of those maps onto a rationale the Panel has actually used to deny transfer. The Panel has repeatedly refused centralization where a limited number of actions, courts, and counsel makes informal cooperation "eminently feasible," and where the efficiencies of § 1407 would be marginal because the common pretrial work is substantially complete. A single-defendant products docket with one plaintiffs' firm at its center is exactly the fact pattern in which those denials issue.

The counterweight is equally familiar. A growing inventory that has already generated filings in 17 districts presents a real risk of duplicative depositions of the same Apple witnesses and inconsistent rulings on the same Daubert Challenge and privilege questions. And the claim that discovery is "largely done" is a claim about the Hughes record — not about the record in 27 cases with different claimants, different devices, different alert histories, and different state-law elements.

Where would the Apple AirTag MDL be located — Arizona or Florida?

If the Panel centralizes, the transferee selection is the day's second decision and, for anyone who will actually litigate the docket, the more consequential one.

Claimants asked for the District of Arizona before Judge Diane Humetewa. Apple's alternative is the Middle District of Florida, on the basis that several of the pending actions are already there. The Panel's transferee criteria are practical rather than doctrinal: where the actions and evidence are concentrated, the transferee court's caseload and docket conditions, the judge's MDL experience, and accessibility.

What turns on the answer is not geography but governance. The transferee judge decides the leadership structure and therefore who controls the common work; the Plaintiff Fact Sheet and any Census Registry that will screen the inventory; the Bellwether Trial pool and selection method; the Common Benefit Fund percentage and, critically, whether the holdback reaches cases resolved outside the MDL; and the scheduling of the general expert fight over whether a consumer Bluetooth tracker can be defectively designed against criminal misuse. Two different judges will run that program two different ways, and the difference is worth more to both sides than the flight time to the courthouse.

The settlement-pace argument cuts both ways

Apple's strongest fact is also its most exposed one. Twenty settled actions in two months is evidence that the parties can resolve cases without a transferee judge. It is equally evidence that the claims have settlement value, that the inventory is maturing, and that a claimant-by-claimant resolution process is under way without any of the structural protections an MDL supplies — no court-supervised Settlement Matrix, no Special Master over allocation, no common benefit accounting, and no transparency for later-filing claimants about what earlier ones received.

That asymmetry is the standard argument for centralization in an inventory that is settling early. Claimants who file in month eighteen have no way to price their case against what was paid in month three. A transferee court is the only mechanism that produces a public benchmark. The Panel has, in other dockets, treated an active but unsupervised settlement program as a reason to centralize rather than a reason to abstain — precisely because the § 1407 inquiry is about the just conduct of the litigation as a whole, not only about speed in the cases that happen to be furthest along.

There is a second-order effect that neither brief will emphasize. Settling 20 actions — by one contemporaneous account about half the cases brought over the tracker — removes from the docket the files most likely to be the best-developed, and a Panel counting live actions on the hearing date sees a smaller inventory than the one the moving claimants described three weeks earlier. Whether that shrinkage reads as "informal coordination is working" or as "the defendant is picking off the count before the vote" is the argument the Panel will hear on Thursday.

Third-party misuse: the defense at the center of the merits

Whatever the Panel does with the forum question, the merits question is the same, and it is not a conventional products question. In a standard Design Defect case, the product injures the user. Here the product performs exactly as designed for the person who bought it, and the injury falls on a non-user whom the purchaser targeted.

That framing is why the defense bar has been watching this docket since long before the MDL motion. A plaintiff's theory that a manufacturer must design against deliberate criminal misuse by a third party, and warn a person who never bought the product, would reach well past Bluetooth trackers — to any consumer device whose lawful function can be turned to surveillance or harm. The doctrinal pressure points are superseding cause, the scope of duty to a non-purchaser, and whether a risk-benefit analysis can treat a product's core feature set as the defect.

The claimants' answer is specific rather than abstract: that Apple's internal materials show it understood its safeguards would deter rather than prevent malicious use; that the volume of stalking reports demonstrates notice at scale; and that the meaningful cross-platform protections — the unwanted-tracking alerts built on the specification Apple and Google shipped in May 2024 — arrived three years after launch, which is a feasible-alternative-design argument dressed in the defendant's own release notes.

What happens after the JPML rules

The Panel typically issues orders within a week or two of argument. Both outcomes have a defined sequence, and neither is dramatic in the week it lands.

Outcome	Immediate steps	Practical consequence
Transfer granted	Panel issues transfer order naming the transferee court and judge; conditional transfer orders follow for later-filed and tag-along actions; transferee court sets an initial case management conference and takes leadership applications	Leadership, fact sheets, a Census Registry, bellwether structure and the common benefit holdback are set within roughly 60 to 120 days; individual settlements continue but against a court-supervised backdrop
Transfer denied	Each action proceeds in its filing district; any coordination is by agreement among counsel; a renewed motion is possible if the inventory grows materially	Duplicative discovery risk returns to each district; settlement remains a private, case-by-case process without a public benchmark; forum-shopping incentives reappear on both sides

Two details matter for anyone tracking the docket. A denial is not a merits ruling and does not bar a later motion on a changed record — the Panel has centralized dockets it previously declined when the action count grew. And a grant does not create a class: there is no opt-out, no classwide judgment, and no automatic compensation. Each claimant still has to prove their own case, and each case is still governed by its own state's law and its own Statute of Limitations.

What it means for plaintiffs' firms, Apple's defense team, and claimants

For plaintiffs' firms holding AirTag inventory. The transferee-selection outcome, not the grant itself, determines the economics. A judge who extends the common benefit assessment to cases resolved outside the MDL changes the return on every case a firm keeps out; a judge who does not, rewards inventory held back. Firms building volume on the assumption that a 2026 MDL will produce a 2027 Global Settlement should price the denial scenario, in which every case is worked and settled individually in its home district.

For Apple's defense team. The opposition is a calculated trade. Defeating centralization preserves a favorable discovery posture and an unsupervised settlement channel, but leaves the company litigating the same corporate-knowledge record in 17 districts and exposed to an outlier ruling anywhere. Winning the alternative — Florida over Arizona — is the hedge, and the brief's structure suggests the hedge is doing real work.

For claimants and their counsel. Nothing about the 24 September argument changes a filing deadline. State limitation periods run on their own schedule, and neither a pending centralization motion nor a transfer order tolls them. A claimant whose case settles before any MDL exists settles without a published benchmark; one who files after a transfer order will be routed into a fact-sheet and registry process before any individual attention is paid to the file.

For clause drafters and corporate defendants watching from outside. This docket is the cleanest current test of whether a consumer-hardware company can be held to design against criminal third-party misuse. A centralized docket with a bellwether program produces that answer in public, on a schedule. A denial produces it in fragments, in whichever district gets to summary judgment first.

Frequently asked questions

What is the Apple AirTag MDL?

It is the proposed federal Multidistrict Litigation docketed as MDL No. 3199, In re: Apple AirTag Marketing, Sales Practices, and Products Liability Litigation — a request filed on 4 August 2026 to centralize the federal AirTag stalking actions before one judge for pretrial proceedings. As of 20 September 2026 the Panel has not issued a transfer order, so the MDL does not yet exist.

When is the JPML hearing on the Apple AirTag lawsuits?

Thursday, 24 September 2026, at the Panel's hearing session in Chicago, Illinois. The same session includes argument on centralizing the multistate cyclosporiasis outbreak actions (MDL No. 3196), the Amazon tariff refund actions (MDL No. 3197), the Abbott and Nevro spinal cord stimulator petitions, and NCAA eligibility antitrust actions.

Who can file an Apple AirTag stalking lawsuit?

The claimants in the pending federal actions are people who allege that another person used an AirTag to track their location without consent and that they suffered injury as a result — reported claims include assault, post-traumatic stress, relocation costs, and security expenses. Several have filed under pseudonyms. Each action is governed by the law of its own state, including that state's Statute of Limitations, and the pending centralization motion does not toll any deadline.

How is an MDL different from a class action?

A class action resolves claims collectively: one judgment or settlement binds every class member who does not opt out. An MDL does not aggregate the claims at all — it aggregates the pretrial work. Each case keeps its own claimant, its own facts, and its own governing law, and under § 1407 each returns to its originating district for trial absent a waiver. That distinction is the reason this litigation exists in its current form: when Class Certification failed in Hughes, the claims did not disappear, they became an inventory of individual suits.

How many Apple AirTag stalking lawsuits have been filed?

The 4 August 2026 centralization motion identified 27 related actions across 17 district courts; contemporaneous reporting described at least 20 actions in at least 17 federal districts. Apple's 25 August brief then reported 20 actions settled in the previous two months, so the number of live federal actions on the hearing date is lower than the number originally identified. No public source gives a verified count as of 20 September 2026.

What happened to the Hughes v. Apple class action?

Hughes v. Apple, No. 3:22-cv-07668-VC (N.D. Cal.), survived dismissal in part on 15 March 2024 as to negligence and strict products liability claims, but Judge Vince Chhabria denied class certification in March 2026, citing state-law variation and the individualized nature of each alleged stalking incident. The claimants refiled individually — at least 16 actions on 2 May 2026 — and those individual filings are the actions now before the Panel.