The Voice Trust Recession: What the AI Voice-Cloning Backlash Means for Your Dealership's Phones

The week the phone's credibility problem went mainstream

On August 28, more than 80 performers - including Bridgerton's Nicola Coughlan, Hugh Bonneville, and Matt Lucas - signed an open letter asking the UK government to make a person's voice part of their statutory rights, backed by a petition to give every citizen legal ownership of their own voice. The campaign, Save Our Voices Now, cites polling that 28% of UK adults report having been targeted by a voice-cloning scam.

An actors' letter in London can feel like a long way from a dealership in Ohio. The polling number is the part that should travel. When roughly one in four adults says a cloned voice has already been used on them personally, consumers stop extending the benefit of the doubt to any voice they were not expecting to hear. And a US dealership's revenue engine - sales follow-up, service reminders, recall outreach, declined-work callbacks - is built almost entirely on voices customers were not expecting to hear.

Myth vs. data: The comfortable explanation for falling contact rates is that customers simply hate phone calls now. The data points at trust, not preference. Pew Research Center found that 80% of US adults generally do not answer cellphone calls from unknown numbers - yet TransUnion's consumer research finds 76% say they would likely answer a call from a business they know if the company's name and logo appeared on screen. Customers have not stopped taking calls. They have stopped taking calls they cannot verify.

Voice went from proof of identity to raw material in about three years

The technical threshold collapsed first. McAfee's "Artificial Imposter" research showed that free, publicly available tools could produce an estimated 85% voice match from about three seconds of audio. In the same study - a survey of 7,054 adults across seven countries - one in four respondents said an AI voice scam had happened to them or someone they know, and 77% of the people it happened to lost money.

The volume followed. Hiya's Global Call Threat Report flagged 12.5 billion suspected spam calls worldwide in the first quarter of 2025, up from 11.3 billion the prior quarter - the equivalent of well over 100 million unwanted calls per day. Deloitte's Center for Financial Services projects that generative AI could push US fraud losses from 12.3 billion dollars in 2023 to as much as 40 billion dollars by 2027, a 32% compound annual growth rate, with voice deepfakes among the named drivers.

Regulators have already responded once. In February 2024, the FCC ruled unanimously that AI-generated voices count as "artificial" under the Telephone Consumer Protection Act - meaning outbound calls using synthetic voice require the same prior consent as any robocall, with statutory damages that can reach 1,500 dollars per willful violation through a private right of action. A growing list of states has layered AI disclosure requirements on top of that federal baseline.

Put those pieces together and the story is consistent: cloning is trivially cheap, fraud volume is compounding, and both regulators and consumers have shifted from "trust by default" to "verify or ignore."

The trust picture, by the numbers

None of these figures were produced with car dealers in mind. That is exactly why they matter: they describe the environment every dealership phone call now enters.

What a trust recession does to dealership phone math

Start with outbound. Most BDC productivity models still assume a version of 2019 physics: more dials produce proportionally more conversations. When four out of five consumers decline unknown numbers on principle, that assumption fails - and pushing harder can make it worse. Carrier analytics engines score calling behavior, and numbers that dial in high volume with low answer rates and short durations are the classic profile that earns a "Spam Likely" label. Legitimate businesses get mistagged this way often enough that an entire remediation industry exists for it. An aggressive dial strategy can quietly destroy the reputation of the very numbers a store depends on.

Now look at inbound. Every force above pushes value toward the call the customer chooses to make. An inbound caller has self-verified: they looked up the store, dialed it, and extended trust first. That makes inbound calls both scarcer and more valuable than they were even two years ago - which raises the cost of the industry's oldest bad habit. We have written before about what a missed call actually costs and about the subtler failure of calls that are answered but never booked. In a low-trust environment, those leaks stop being an efficiency problem and start being a strategy problem: the store is squandering the one channel the trust recession has not degraded.

There is also a defensive dimension. The same cloning tools aimed at consumers are being aimed at businesses, including dealership phone rooms handling payoff, wire, and identity requests. A store's callers doubt the store; the store now has reason to doubt some of its callers. Trust has to be engineered in both directions.

A legitimacy playbook for the phone-first store

The stores that win phone traffic in this environment will be the ones that treat trust as infrastructure rather than assuming it. The practical moves are not exotic.

1. Treat number reputation as a balance-sheet asset. Consolidate outbound calling onto a small set of consistent, properly registered numbers with full STIR/SHAKEN attestation through your phone provider, and check monthly whether your numbers carry spam labels on major carriers. Branded calling - displaying the store's name and logo on the recipient's screen - is the highest-leverage fix available; TransUnion reports large answer-rate lifts when calls carry verified branding, including a 105% lift in one published public-sector case study (vendor-reported, but directionally consistent with its finding that 76% of consumers would answer a recognized business).
2. Rebalance effort from outbound chase to inbound excellence. If outbound connect rates are structurally capped by consumer behavior, the marginal dollar moves to answering every inbound call - nights, weekends, service-lane peak hours - and converting it to a booked appointment rather than a message slip. The trust recession does not touch calls the customer initiates; failing to answer them does.
3. Disclose AI plainly, because disclosure is now a trust marker. The FCC's ruling covers consent for outbound AI voice; several states require disclosure regardless of direction. But compliance is the floor. A store whose phone agent - human or AI - identifies itself clearly, states its purpose, and never asks for sensitive data by phone is building exactly the verification signal consumers are scanning for. Evasiveness reads as fraud now, even when it is just awkward software.
4. Give customers a verification path. Publish one primary number, use it consistently for calls and texts, and train every script - again, human or AI - to invite the customer to call the main store line back if anything feels off. Businesses that make verification easy convert skeptics; businesses that resist it lose them.
5. Harden the store's own phone room. Written callback-verification procedures for wire instructions, payoff changes, and one-time-password requests cost nothing and defeat most voice-clone pretexting attempts.

The direction of travel

The UK petition for statutory voice ownership will not bind a US dealership, but it marks where public sentiment is heading, and US law tends to follow the same arc: Denmark has moved toward personal rights over face and voice, the FCC has already classified synthetic voices as regulated artificial calls, and state disclosure statutes keep multiplying. The plausible end state is a phone network where legitimate calls are verified, branded, and disclosed - and everything else is presumed hostile. Stores that build for that world early will not experience the transition as a compliance burden. They will experience it as answer rates their competitors cannot explain.

The phone is not dying. Unverified calling is dying. For dealerships, which still transact more revenue by phone than any other channel, that distinction is the whole game.

If your store is rebalancing toward inbound capture, that is the side of the ledger Carbuki works on: AI voice agents that answer every inbound call 24/7, identify themselves honestly, and book the appointment while trust is highest. Details at carbuki.com.

Sources

• Variety, "Nicola Coughlan, Matt Lucas, Hugh Bonneville and More Back Campaign Against AI Voice Cloning" (Aug 28, 2026): https://variety.com/2026/digital/global/nicola-coughlan-matt-lucas-ai-voice-cloning-law-1236845817/
• ITV News, "Nicola Coughlan and Hugh Bonneville among stars backing campaign against AI voice cloning" (Aug 28, 2026): https://www.itv.com/news/2026-08-28/nicola-coughlan-and-matt-lucas-among-stars-backing-ai-voice-cloning-campaign
• Pew Research Center, "Most Americans don't answer cellphone calls from unknown numbers" (2020): https://www.pewresearch.org/short-reads/2020/12/14/most-americans-dont-answer-cellphone-calls-from-unknown-numbers/
• TNS via Business Wire, "TNS Survey: 75% of Americans Will Never Answer Calls from Unknown Numbers" (2022): https://www.businesswire.com/news/home/20220726005226/en/
• McAfee, "Artificial Imposters - Cybercriminals Turn to AI Voice Cloning for a New Breed of Scam" (2023): https://www.mcafee.com/blogs/privacy-identity-protection/artificial-imposters-cybercriminals-turn-to-ai-voice-cloning-for-a-new-breed-of-scam/
• Hiya, "Global Call Threat Report" (Q1 2025): https://blog.hiya.com/global-call-threat-report-q1-brings-a-new-assortment-of-phone-scams
• Deloitte Center for Financial Services, "Generative AI is expected to magnify the risk of deepfakes and other fraud in banking" (2024): https://www.deloitte.com/us/en/insights/industry/financial-services/deepfake-banking-fraud-risk-on-the-rise.html
• Federal Communications Commission, "FCC Makes AI-Generated Voices in Robocalls Illegal" (Feb 2024): https://www.fcc.gov/document/fcc-makes-ai-generated-voices-robocalls-illegal
• TransUnion, "What Is Branded Calling?" (consumer research; vendor case study): https://www.transunion.com/faq/what-is-branded-calling