Why AI Spam Calls Are Getting Worse — Not Better

If you feel like spam calls are getting smarter, you're not imagining it. The numbers show the problem is getting measurably worse — and artificial intelligence is a big part of the reason why.

U.S. consumers received 52.5 billion robocalls in 2025, barely a 1% drop from the year before. But the mix shifted in a troubling direction: telemarketing and scam calls jumped 15.4%, rising from 49% to 57% of all robocalls. Meanwhile, legitimate calls like payment reminders and appointment notifications actually declined.

The human toll is just as stark. According to U.S. PIRG Education Fund research, 31% of American adults receive at least one scam call every day, and 21% get several. Victims of scam calls lost an average of $3,690 in the first half of 2025 — a 16% increase year over year.

Why AI makes it worse

AI voice technology has removed the biggest tell of a scam call: the robotic, stilted delivery. Synthetic voices now pause naturally, respond to questions, and mimic local accents, making fraud far more believable and easier to execute at scale. The FCC has clarified that AI-generated voice calls are illegal without prior express consent — but the rule only helps if it's enforced.

Why regulation alone hasn't fixed it

Enforcement is ramping up, yet gaps remain. In August 2025, the FCC shut down 1,388 non-compliant phone companies, and all 51 state attorneys general launched a joint crackdown on robocall traffic. Still:

• Only 44% of phone companies have fully implemented STIR/SHAKEN caller-ID verification — down from 47% in 2024
• 32% of providers reported no implementation at all
• Scammers ignore the Do Not Call Registry entirely, since it only binds law-abiding telemarketers

That last point matters most. As consumer guidance puts it, the registry is your first layer of defense — not your last.

This is why legitimate operators matter to the equation. Services like My AI Call Center run structured campaigns only against approved, permissioned, or reviewed contact lists, honoring opt-outs and DNC requests immediately — proof that AI calling and consent discipline can coexist. The rest of this guide covers the practical steps you can take to shut down the calls that don't play by those rules.

Your Legal Rights: Opt-Outs, DNC Registry, and the TCPA

You don't have to just hang up and hope. Federal rules give you real, enforceable rights against AI robocalls — and knowing them changes how you handle the next call that comes in.

The strongest protection comes from the TCPA. The FCC has clarified that AI-generated voice calls are illegal unless you have agreed to receive them or the caller is exempt. That means an AI voice call without your prior express consent isn't just annoying — it's a violation of federal rules.

Even if you did consent at some point, you can take it back. The FCC states plainly that you may opt out of any robocall or robotext at any time and in any reasonable manner, even if you previously gave consent. Telemarketers using prerecorded calls must offer an opt-out option at the start of the call, and any do-not-call request made during a call must be honored immediately. Saying "stop calling me" counts.

Keyword opt-outs make this even simpler on AI-driven calls. Well-run campaigns — like those My AI Call Center operates — treat words like STOP and REVOKE as immediate, binding opt-outs, logged and honored on the spot rather than requiring a callback or email. If a caller won't accept your opt-out, that's a red flag in itself.

The National Do Not Call Registry is your baseline defense. Registration is free at donotcall.gov or by calling 1-888-382-1222 from the number you want to list. The registry has grown to over 258 million active registrations as of the FTC's 2026 Do Not Call Report, and legitimate telemarketers must stop calling within 31 days of your registration.

Know the registry's limits, though. As consumer guidance points out, it works well against companies that follow the law — but scammers, political organizations, charities, and survey callers are either exempt or simply ignore it. Treat it as your first layer of defense, not your last.

To put your rights to work:

• Register your number on the National Do Not Call Registry at donotcall.gov.
• Say "stop calling" during any live call — the request must be honored immediately.
• Text STOP or REVOKE to end consent on any AI or automated campaign.
• Report violations to the FCC at consumercomplaints.fcc.gov.

Reporting matters more than most people realize. The FCC has issued hundreds of millions of dollars in enforcement actions against illegal robocallers, and in August 2025 it shut down 1,388 non-compliant phone companies for failing to file robocall mitigation data. Complaints feed that enforcement pipeline.

Responsible operators make this easy on their end, too. Before any campaign launches, My AI Call Center reviews list sources and consent records, and declines bought lists without clear permission — so the calls you receive are ones you actually agreed to. When consent, opt-outs, and DNC registration work together, the unwanted calls have nowhere to go.

Build a Layered Defense: Registry, Carrier, and Device Blocking

If you're waiting for one magic button that makes AI spam calls disappear, you'll be waiting a long time. As one consumer privacy guide puts it plainly, no single step will block 100% of robocalls — but combining a few simple tools can cut the volume down to almost nothing. The good news: building those layers takes less than an hour.

Start with the Do Not Call Registry. Register at donotcall.gov or call 1-888-382-1222 from the number you want protected. Legitimate telemarketers must stop calling within 31 days of registration, and the registry had over 258 million active registrations as of the FTC's 2026 report. But treat it as your first layer, not your last — scammers, political organizations, and charities either ignore the list or are exempt from it.

Next, turn on the free blocking tools your carrier already provides. The FCC's call-blocking guidance points to AT&T ActiveArmor, T-Mobile ScamShield, and Verizon Call Filter as first-line defenses that flag or filter suspicious calls before they ever ring your phone. These tools are typically free and take minutes to activate.

Then add the features built into your device. Your blocking stack should look something like this:

• iOS: enable "Silence Unknown Callers" so numbers not in your contacts go straight to voicemail
• Android/Google Pixel: use "Call Screen" to let Google's assistant answer and transcribe suspected spam calls
• Carrier apps: activate ActiveArmor, ScamShield, or Call Filter depending on your provider
• Third-party apps: consider Hiya, Nomorobo, or YouMail for an extra filtering layer

One caution on that last layer: security reporting has found that many top robocall-blocking apps share your phone number with analytics firms and upload device information to companies like Facebook without explicit consent. Read the privacy policy before installing anything.

The stakes justify the effort. U.S. consumers received 52.5 billion robocalls in 2025, with roughly 200 robocalls per adult — and scam victims lost an average of $3,690 each in the first half of 2025 alone, according to FTC data.

The same principle applies on the sending side. At My AI Call Center, every campaign runs only against approved, permissioned, or reviewed lists, with DNC requests honored immediately and logged — because responsible calling and effective blocking are two halves of the same system.

Cut Off the Source: Protect Your Number and Report Violations

The most effective way to stop AI spam calls is to make your number harder to find in the first place. Scammers and non-compliant telemarketers pull contact data from public records, data broker sites, and leaked databases — fueling the 29.6 billion unwanted telemarketing and scam calls Americans received in 2025, a 15.4% jump from the prior year. Industry data shows this category now makes up 57% of all robocalls, up from 49% in 2024.

Start by removing your number from major people-search sites like WhitePages, Spokeo, and BeenVerified. Use a virtual or secondary number for online signups, loyalty programs, and any non-essential account creation — this keeps your primary line out of marketing databases that get resold or scraped. At My AI Call Center, we see this principle in reverse every day: campaigns only run against approved, permissioned, or reviewed lists, and bought lists without clear consent records are declined before a single call is placed.

When violations happen, report them. The FCC and FTC rely on consumer complaints to target enforcement. In August 2025 alone, the FCC shut down 1,388 non-compliant phone companies for failing to file proper robocall mitigation data, and all 51 U.S. attorneys general launched Operation Robocall Roundup, sending warning letters to 37 voice providers. Filing a complaint at consumercomplaints.fcc.gov or the FTC's reporting portal adds your data to the enforcement pipeline.

• Opt your number out of data broker and people-search sites
• Use a virtual number for online forms, shopping, and non-critical accounts
• Report unwanted AI calls to the FCC and FTC with date, time, and caller details
• Exercise your opt-out rights during any prerecorded call — they must honor it immediately

The FCC confirms you may opt out of any robocall or robotext at any time, in any reasonable manner, even if you previously gave consent. Combining source reduction with active reporting creates pressure on the ecosystem that enables AI spam — and supports the same compliance standards legitimate outbound campaigns are built on.

The Other Side of the Problem: Compliant AI Calling Exists

Here's the thing most spam-call articles won't tell you: the same rules that protect consumers also define how legitimate AI calling works. The problem isn't AI — it's callers who skip consent, disclosure, and opt-out handling.

The regulatory baseline is clear. The FCC has ruled that AI-generated voice calls are illegal unless the consumer agreed to receive them or the caller qualifies for an exemption, and consumers may opt out of any robocall at any time in any reasonable manner, even if they previously gave consent. Compliant callers treat those rules as campaign requirements, not suggestions.

That compliance posture looks like this in practice:

• Calling only approved, permissioned, or reviewed contact lists — never indiscriminate dialing
• AI disclosure on every call, with recipients able to ask if the call is AI-assisted, request a human, or opt out
• Opt-out requests honored immediately and logged
• Do-not-call requests carried across all campaigns into DNC records
• Calls run only in approved windows, respecting quiet hours and state-specific rules

The stakes for ignoring this are rising. The FCC shut down 1,388 phone companies in August 2025 for failing to properly file robocall mitigation data, and imposed a $10,000 penalty for false or inaccurate filings. Meanwhile, scam and telemarketing calls surged 15.4% in 2025 to 29.6 billion calls — 57% of all robocalls — which means the gap between compliant and non-compliant callers is widening fast.

This is where list discipline matters most. The National Do Not Call Registry had over 258 million active registrations, and legitimate telemarketers must stop calling registered numbers within 31 days. A caller who can't verify where a list came from — or whether the people on it gave permission — has no business dialing it.

My AI Call Center runs structured campaigns against consent-checked lists only. Before any campaign launches, list source and consent records are reviewed; bought lists without clear permission records are flagged and, in most cases, declined. The first campaign review is free, so list and consent issues surface before anything is spent.

The takeaway cuts both ways. If you're a consumer, the compliance framework gives you real rights — register, opt out, report. If you're a business that wants AI calling to work for you, the same framework is your roadmap: one clear goal per campaign, a permissioned list, disclosure on every call, and opt-outs honored without exception.