The Autodialer Question That Keeps Causing Lawsuits

If you run outbound campaigns, the legal definition of an autodialer is not a theoretical debate. It is a liability risk that lands on your balance sheet. The Telephone Consumer Protection Act (TCPA) has long been the source of confusion for businesses using automated dialing systems.

The confusion stemmed from a split among six federal circuit courts. Three circuits interpreted the law narrowly, while three interpreted it broadly. This divergence created a legal minefield for companies dialing from pre-established lists. You could face a lawsuit in one state and be compliant in another.

The U.S. Supreme Court ended this split in its unanimous 2021 decision in Facebook, Inc. v. Duguid. The Court held that a device qualifies as an autodialer only if it has the capacity to store or produce numbers using a random or sequential number generator. This ruling clarified that equipment which simply stores and dials pre-established lists does not meet the federal definition of an autodialer.

This clarity does not eliminate legal exposure. Statutory damages under the TCPA range from $500 to $1,500 per call. A single campaign with thousands of contacts can generate millions in potential liability. The risk is compounded by the fact that the TCPA separately prohibits calls using an artificial or prerecorded voice.

For businesses using AI-generated voices, this distinction is critical. Your dialing method may no longer trigger the autodialer definition, but your voice technology does. Since 2012, telemarketers must obtain prior express written consent before making calls with an artificial voice. The Supreme Court’s decision does not change this consent requirement. It only narrows the scope of what constitutes an autodialer.

The legal landscape has shifted, but the compliance burden remains. You cannot assume that because your dialing system is not an autodialer, you are free from TCPA restrictions. The separate prohibition on artificial voices still applies. You still need consent.

Consider the practical implications for your operations:

• Your dialing technology likely falls outside the narrow autodialer definition if it uses pre-loaded lists.
• Your use of AI voices triggers the separate artificial voice consent requirement.
• Statutory damages of up to $1,500 per call still apply to violations of the artificial voice provision.
• Prior express written consent is mandatory for any call using an artificial voice, regardless of dialing method.

My AI Call Center operates within this framework by treating AI voices as artificial voices under the TCPA. We require prior express consent for all campaigns. We review list sources and consent records before launch. This discipline protects you from the most common legal pitfalls.

The Supreme Court’s ruling was a win for legitimate businesses. It stems the tide of class action suits against companies that dial from approved lists. But it does not offer a free pass. The consent requirements remain strict. The damages remain high. The need for documented permission remains unchanged.

Guessing at the answer is no longer an option. The law is clear on what constitutes an autodialer. It is equally clear on what constitutes an artificial voice call. Your compliance strategy must address both. You need to know exactly how your technology works and what consent you have on file.

The cost of ignorance is measured in dollars, not just legal fees. A single violation can trigger a lawsuit. A pattern of violations can trigger a class action. The difference between a compliant campaign and a lawsuit often comes down to one detail: the consent record.

You do not need to navigate this complexity alone. But you do need to navigate it with precision. The rules are specific. The penalties are severe. The definition is now clear. Your next step is to ensure your consent records match that clarity.

What the Supreme Court Actually Decided in Facebook v. Duguid

For years, businesses couldn't agree on what made a dialer an "autodialer" under federal law. On April 1, 2021, the Supreme Court settled it, unanimously, in Facebook, Inc. v. Duguid.

The holding is simple to state. Writing for a 9-0 Court, Justice Sotomayor held that a device qualifies as an "automatic telephone dialing system" only if it has the capacity to either store or produce telephone numbers using a random or sequential number generator. Equipment that merely stores a list of numbers and dials them automatically doesn't qualify, no matter how fast it dials (Wiley Law).

That reading rejected the Ninth Circuit's test head-on. The lower court had held that an autodialer need only "store numbers to be called" and "dial such numbers automatically," reading the random-or-sequential phrase as applying only to "produce." The Supreme Court sided with the narrower circuits and reversed (Crowell & Moring).

The case itself started small. In 2014, Noel Duguid received Facebook security-alert texts even though he had no Facebook account. A California district court dismissed his TCPA suit, the Ninth Circuit reversed, and the Supreme Court then reversed the Ninth Circuit. More than 20 parties filed amicus briefs along the way, a sign of how much hung on the definition.

To understand why the Court's answer mattered, it helps to look at what came before:

• 1991: Congress enacted the TCPA, and the autodialer definition has remained unchanged since then.
• 2015: The FCC's Declaratory Ruling defined "capacity" to include "potential functionalities" and "future possibility," an extremely broad reading.
• 2018: The D.C. Circuit invalidated that reading in ACA International v. FCC, calling its reach an "eye-popping sweep" that could render every smartphone an ATDS.
• 2021: The Supreme Court adopted the narrow, text-based interpretation.

The Court's reasoning leaned on ordinary grammar. The comma in the statutory definition signaled that "using a random or sequential number generator" modifies both "store" and "produce." A broader reading, Sotomayor wrote, "would take a chainsaw to these nuanced problems when Congress meant to use a scalpel," potentially capturing ordinary cell phone owners in commonplace usage. Congress had targeted equipment that risks dialing emergency lines randomly or tying up sequentially numbered lines at a single entity.

The practical takeaway: most businesses dial from pre-established lists rather than generating numbers, so the ruling may stem a wave of TCPA class actions carrying $500 to $1,500 in statutory damages per call. But the decision doesn't erase compliance duties. The TCPA separately prohibits artificial or prerecorded voice calls, and since 2012 telemarketers have needed prior express written consent for those calls. That's why My AI Call Center treats its AI-generated voices as artificial voices under the TCPA and requires consent records before any campaign launches, regardless of how the autodialer definition is drawn.

Why the Narrow Definition Doesn't Clear Your Consent Obligations

Here's the catch, and it's a big one. If you run AI-generated voice calls, the narrow autodialer definition doesn't get you out of the consent conversation at all.

The Supreme Court itself flagged this in Facebook v. Duguid. Justice Sotomayor noted that the TCPA separately prohibits using "an artificial or prerecorded voice" on various types of phone lines. That prohibition survived the ruling untouched. So even if your dialer stores and dials a pre-approved list with no random or sequential number generation, the voice on the call still triggers its own set of rules.

AI-generated voices fall squarely into the artificial-voice category. And since 2012, the rules have been strict: the FCC requires prior express written consent before making telemarketing calls that use an autodialer or contain a prerecorded or artificial voice. That means an agreement in writing, bearing the called person's signature, that clearly authorizes the seller to deliver telemarketing messages using that technology.

The old shortcut is gone, too. The 2012 revisions eliminated the "established business relationship" exception for consent, so you can no longer lean on the fact that someone is your customer to justify an artificial-voice marketing call. If you've been assuming your existing customer list is fair game, that assumption is a liability.

Why does this matter so much in practice? Statutory damages run $500 to $1,500 per call, and the TCPA gives consumers a private right of action to recover up to $1,500 per violation. A sloppy list doesn't just risk a warning; it multiplies fast.

This is why list discipline and consent records are the real compliance work. Before any campaign launches, you need to be able to answer:

• Where did this list come from, and who permissioned it?
• Do you hold written consent records for each contact, and can you produce them?
• Are numbers screened against the national and company-specific do-not-call lists?
• Are opt-outs logged immediately and honored across every future campaign?

At My AI Call Center, we treat this review as the gate, not the afterthought. We check list source, consent records, and calling windows before anything launches, and we tell you plainly if a list won't support the campaign before you spend anything. Legal analysts at Crowell & Moring reach the same practical conclusion: companies should continue to obtain consent where practicable, even after Duguid narrowed the autodialer definition.

The narrow definition changed one question. It didn't change your obligations.

What This Means for Your Next Outbound Campaign

A unanimous Supreme Court decision in April 2021 narrowed what counts as an autodialer, but that ruling did not eliminate your compliance obligations. The Court held that equipment must use a random or sequential number generator to store or produce phone numbers before it qualifies as an ATDS. Most campaigns that dial from pre-established lists fall outside that definition, yet the separate prohibition on artificial or prerecorded voice calls remains fully intact, and that is where the real exposure sits for any AI-voice operation.

For your next outbound campaign, the practical steps line up clearly.

• Keep written consent on file for every contact on your list, signed before any call goes out, and store it where your team can pull it on demand.
• Document list source and calling windows before launch, including the relationship that earned the contact a spot on the list in the first place.
• Honor opt-outs and DNC requests across every campaign you run, and carry those flags into your master suppression list so they never get re-dialed.
• Watch the unresolved "capacity" question, since future FCC and court decisions could reshape how dialing technology gets classified.

The stakes are concrete. Statutory damages run $500 to $1,500 per call, and class actions have piled up around dialing practices that looked routine until someone complained. Even with the narrower ATDS definition, legal analysts at Crowell & Moring recommend that companies continue to obtain consent wherever practicable and follow established best practices when calling or texting individuals.

That is the posture a compliance-forward managed service takes. At My AI Call Center, every campaign starts with a list and consent review before launch, and the team flags or declines bought lists without clear permission records. Opt-outs get logged and honored immediately, and DNC requests carry across all current and future campaigns. The first campaign review is free, and the full cost is quoted before anything dials. If you want to scope a campaign against a list you already trust, plan your campaign and the team will tell you plainly whether the list will support the goal before you spend anything.