The FTC did not create a new law for AI calling. It extended existing laws to explicitly cover it, which is a more significant development for outbound sales teams than a new statute would have been.

The Telemarketing Sales Rule has governed outbound sales calls since 1995. For decades, most of its provisions applied only to consumer calls. The March 2024 update changed two things that directly affect any company running outbound campaigns with AI.

1. The FTC explicitly confirmed that the TSR’s prohibition on robocalls and prerecorded messages applies to calls using AI-generated voices, including neural text-to-speech and voice cloning technology. There was no previous ambiguity about intent here, but the explicit confirmation removed any possibility of arguing that an AI voice occupied a legal grey area.
2. The update extended TSR protections to business-to-business calls. Previously, most B2B telemarketing was exempt from federal scrutiny. That exemption is now significantly narrower. Any B2B call that involves misrepresentation, deceptive claims, or failure to honor opt-out requests now falls under federal enforcement authority.

One month before the FTC update, in February 2024, the FCC issued its own declaratory ruling, FCC-24-17, confirming that AI-generated voices qualify as “artificial or prerecorded voice” under the TCPA. That ruling brought AI calling squarely into the prior express written consent framework that applies to all prerecorded consumer calls.

The two rulings work together. The FCC ruling determines that consent is required before an AI calling campaign can legally reach consumer numbers. The FTC ruling governs how the call must be conducted once it connects, including disclosure requirements and DNC obligations. A single non-compliant AI call can now trigger liability under both, through two separate enforcement channels and two separate penalty structures.

Summary

• The FTC updated its Telemarketing Sales Rule in 2024 to explicitly cover AI-generated voices and expanded it to include business-to-business calls. AI calling without proper consent and disclosure now violates federal law and carries penalties of up to $51,744 per call.
• In February 2024, the FCC made a declaratory ruling that AI voices are considered “artificial or prerecorded voice” under the TCPA, which means that consumer calls must have full prior express written consent.
• These two rulings work together. A single non-compliant AI call can now violate both the TSR and the TCPA simultaneously, creating dual liability.
• The enforcement record shows the FTC acts on these cases. Outbound sales teams that treat compliance as a training issue rather than a systems issue carry unnecessary liability at every dial.
• A fully managed AI calling platform enforces every federal and state compliance requirement at the infrastructure level before a call is placed, removing the liability from human execution.

The Enforcement Record Shows the FTC Acts

The FTC’s action against an AI calling platform in 2025 was not a warning shot. It was a completed enforcement case, and the outcome should inform how every outbound team evaluates its compliance posture.

The FTC filed its complaint in the US District Court for the District of Arizona in August 2025. The agency alleged that the platform and its operators made false earnings claims to buyers, misrepresented the performance and capabilities of their services, and violated the Telemarketing Sales Rule by failing to provide required disclosures and earnings claim statements to purchasers.

The case reached a settlement that effectively ended the platform’s operations. Thousands of outbound teams that had been using or evaluating that platform suddenly needed to find an alternative, and many discovered for the first time that their vendor’s legal compliance was not guaranteed.

Three lessons apply directly to any team using AI for outbound calling.

The FTC treats platforms and operators as directly liable. The enforcement action targeted the platform’s founders and operators personally, not just the corporate entity. Clients who ran non-compliant campaigns also carried liability for the calls their campaigns made.

Performance claims about AI calling are under regulatory scrutiny. Any platform that makes earnings guarantees, income projections, or ROI claims without verifiable data is operating in territory the FTC now treats as high-risk. A platform’s marketing claims are part of its compliance profile.

Vendor compliance is not the same as operational compliance. A platform can build technically legal infrastructure and still expose its clients to liability if the client’s lead lists, consent records, or campaign settings do not meet the required standards.

What Your Outbound Team Must Do Right Now

Outbound teams running AI calling in 2026 carry three specific federal compliance obligations that did not exist in the same form five years ago, and state law adds additional requirements on top of each one.

Consent Must Be Prior, Express, Written, and One-to-One

The prior express written consent requirement under the TCPA means you must have documented proof that a consumer specifically agreed to receive AI-powered or automated calls from your organization before your first dial. Verbal opt-in is not sufficient for consumer cell phones. A general website form without a clear consent checkbox does not satisfy the requirement.

The one-to-one consent principle, affirmed by FTC guidance and supported by FCC rulemaking intent, requires that the consent specifically name your company. A consumer who consented to calls from a lead generation aggregator did not give consent to calls from your sales team. Businesses must honor opt-out requests within 10 business days, reduced from the previous 30-day window, and that obligation applies across all channels simultaneously. An opt-out recorded on a voice call must suppress the number in your email, SMS, and calling campaigns immediately.

DNC Validation Must Happen Before Every Dial

The National Do Not Call Registry requires scrubbing every number before each campaign, rather than on a monthly batch schedule. Telemarketers must verify the registry before calling and scrub contact lists against it at least every 31 days, but for high-volume AI calling operations, every 31 days is not sufficient protection. A number added to the registry after your last scrub and before your next campaign could generate a live TCPA violation with each dial.

State-Specific Dialing Windows Apply Based on the Recipient’s Location

Federal law prohibits calling before 8 AM or after 9 PM in the recipient’s local time zone. At least 15 states impose tighter windows. Oregon and Florida cap calls at 8 PM. Rhode Island ends calls at 6 PM on weekdays. Maine prohibits Sunday calling. For a national outbound campaign, the only safe default is to apply the most restrictive state window across the entire list and expand only when you have confirmed location and state-specific compliance reviews.

The Cost Math When You Get This Wrong

The $51,744 per call FTC penalty is not the number most outbound teams worry about. The private right of action under the TCPA is what actually generates the litigation volume that reshapes how regulated industries approach outbound compliance.

TCPA class actions reached a record in Q1 2026, with 220 class actions filed in March alone. The structure of TCPA litigation makes high-volume AI calling campaigns a particularly attractive target. Because the statute allows $500 to $1,500 per call, a campaign that places 100,000 calls with a consent or DNC defect can generate $50 million to $150 million in statutory exposure before any negotiation begins. Class counsel does not need to prove harm. The statute is strict liability for willful violations.

Consider a situation in which a solar sales team acquires 50,000 leads from a third-party aggregator and runs an AI calling campaign without verifying that the lead consent specifically names their company. If 10 percent of those leads challenge the consent, that is 5,000 potential TCPA plaintiffs, each with a $1,500 per call statutory claim for every call your campaign placed. The math compounds quickly.

The teams that face the largest exposure are not the ones ignoring compliance entirely. They are the ones who believe their vendor handles compliance when the vendor’s contract explicitly limits their liability to the software layer.

What a Fully Managed Platform Enforces at the System Level

The practical difference between an AI calling API and a fully managed AI calling platform is the location of compliance responsibility. An API puts it on you. A fully managed platform builds it into every dial.

A fully managed platform runs TCPA compliance checks before a call is placed, not after an audit reveals a problem. That means TrustedForm consent tokens are validated at the lead level. Numbers are checked against the National DNC Registry in real time. State-specific dialing windows are enforced based on the recipient’s area code, not the call center’s time zone. Velocity caps prevent the same number from being called more frequently than the applicable rules allow.

When a prospect says “take me off your list” or “stop calling me,” a managed platform detects that opt-out in natural language and immediately suppresses the number across all active campaigns. The suppression is logged with a timestamp and stored as part of the compliance audit trail. No human agent needs to remember to click anything.

Every call result, transcript, recording, and disposition is pushed automatically to your CRM after the call. If an FTC or TCPA inquiry ever arrives, the audit trail exists and is complete.

That is the operational difference between treating compliance as a training problem and treating it as a systems engineering problem. One creates liability at scale. The other removes it.

If your outbound team is grinding through low connect rates and burning through reps, Bigly Sales gives you a better way. Our AI voice agents qualify your leads, book appointments, and hand off warm prospects to your closers so your team spends every hour on real selling.

See what Bigly Sales can do for your pipeline at biglysales.com.

ABOUT BIGLY SALES

Bigly Sales is an AI-powered outbound calling platform designed for sales teams that need to move faster, stay TCPA compliant, and scale without adding headcount. From insurance and mortgage to debt relief and solar, Bigly Sales helps high-velocity teams automate prospecting, qualify leads, and book more meetings with AI voice agents. Learn more at biglysales.com.

FAQs

Does the FTC ruling apply to B2B outbound sales calls made with AI?

The March 2024 TSR update extended the FTC’s jurisdiction to business-to-business telemarketing, removing the broad B2B exemption that previously applied. Calls to business recipients that involve misrepresentation, deceptive claims, failure to disclose the AI nature of the call, or failure to honor opt-out requests now fall under federal enforcement authority. Note that the DNC Registry exemption for verified business landlines still applies in limited circumstances, but that exemption breaks down for calls to personal cell phones used for business purposes.

Can you use AI calling legally for outbound sales in 2026?

Yes, when the operation meets the required consent, disclosure, and suppression standards. You need prior express written consent for campaigns targeting consumers on cell phones. You must identify the AI nature of the call at the outset of each conversation. You must validate numbers against the National DNC Registry before each campaign. You must honor opt-out requests across all channels within 10 business days. Platforms that enforce these requirements at the infrastructure level can run compliant, high-volume AI calling campaigns without generating the litigation exposure that unmanaged operations carry.

What counts as prior express written consent for AI calling?

Prior express written consent requires a signed or digitally agreed record showing that a specific consumer has agreed to receive automated or AI-generated calls from a specifically named seller. The consent must be unambiguous, must not be buried in general terms and conditions, and must specifically authorize the type of call being made. Consent obtained through a third-party lead generation form that does not name your company does not satisfy this requirement under current FTC and FCC guidance.

What is the time frame for honoring a DNC opt-out request?

Federal law requires honoring opt-out requests within 10 business days of receipt. That timeline was reduced from the previous 30-day window. The opt-out must be applied across all channels and all campaigns, not just the one that recorded the request. Platforms that run real-time suppression apply the opt-out immediately rather than waiting for the 10-day window to expire, which eliminates the risk of placing a second call to an opted-out number during that window.

What disclosures must an AI voice agent make at the start of a call?

The FTC’s TSR requires that telemarketers promptly identify the seller’s name and the purpose of the call before making any sales pitch. The FCC’s position on AI-generated voices adds a practical transparency requirement around the automated nature of the call. A compliant AI voice agent identifies who is calling, the purpose of the call, and that the caller is an automated system before moving into any qualifying questions or sales content. This disclosure must be made in the same language as the rest of the call and at a pace and volume that a typical consumer can understand.