Tech

Caller ID Authentication: What STIR/SHAKEN Does and Does Not Do

Federal rules require phone companies to verify caller ID with a system called STIR/SHAKEN. Here is how it works, what the FCC's recent rules changed, and why scam calls still get through.

Green mountains under white clouds and a blue sky

When a call shows a familiar number on your screen, a technical system now helps decide whether that number can be trusted. This article explains how that system, known as STIR/SHAKEN, works, what the Federal Communications Commission (FCC) has required of phone companies, and where the protection stops.

Why caller ID was easy to fake

For decades, the number displayed on a phone screen was simply information the calling network passed along. Nothing in the older telephone system checked whether the caller had any right to use it. That gap made “spoofing” possible: displaying a number other than the one actually placing the call. Scammers use it to imitate a local number, a bank or a government office, because people are more likely to answer a call that looks familiar.

The industry’s answer was a pair of technical standards, STIR (Secure Telephone Identity Revisited) and SHAKEN (Signature-based Handling of Asserted information using toKENs), usually written together as STIR/SHAKEN. According to the FCC’s 2023 order, STIR/SHAKEN is a set of technical standards for authenticating and verifying caller ID information. The FCC describes two parts: the technical process of authenticating caller ID, and a governance process that maintains trust in the certificates behind that information.

How does STIR/SHAKEN work?

The idea resembles a tamper-evident seal. When a call enters the network, the provider that originates it adds a digital signature, created with a certificate the provider holds, that vouches for the caller ID information. That signature travels with the call through the networks between caller and recipient. The provider that delivers the call to you can then check the signature, and if it does not validate, treat the call as suspect.

The system depends on a few pieces:

  • Digital certificates. A provider must obtain a Service Provider Code token and a digital certificate before it can sign calls. According to the FCC’s 2025 rule, all providers with a STIR/SHAKEN obligation must have these.
  • A governance structure. Certificates are issued within a managed system so that carriers can trust one another’s signatures.
  • Attestation. The signing provider records how confident it is in the caller’s identity and right to use the number. The FCC’s 2025 rule says the higher-confidence levels, A and B, may be applied only by the originating provider after it authenticates a call it places onto the public network, and both require a direct authenticated relationship with the customer. The top level, A, additionally requires confirming the customer’s right to use the number shown.

The result is information that can be used by carriers’ analytics tools and, in some cases, by the app or phone that shows a “verified” or “potential spam” label. Whether you see any label depends on your carrier and device.

What the FCC’s 2023 order added

Early rules applied mainly to carriers that placed calls onto the network. In June 2023, the FCC published its Sixth Report and Order in the Federal Register (FCC 23-18), extending obligations to more parts of the call path. According to that document:

  • The first non-gateway intermediate provider that receives an unauthenticated call directly from an originating provider must authenticate it, with a compliance date of December 31, 2023.
  • All providers must take “reasonable steps” to mitigate illegal robocalls, without the FCC prescribing specific measures.
  • All providers must file a certification and a mitigation plan in the Robocall Mitigation Database, a public FCC registry.
  • Providers must block traffic from non-gateway intermediate providers that are not listed in that database.
  • Enforcement tools include per-call forfeitures for blocking failures, with a $2,500 base amount and a $23,727 maximum per call for non-common carriers, removal from the database, and potential revocation of section 214 authority, the legal authorization to operate as a telecommunications carrier.

An “intermediate provider” is a company that carries a call between the originating carrier and the carrier that delivers it. A “gateway” provider is one that brings calls into the United States from abroad.

What changed in 2025: who may sign calls

A later FCC order, adopted November 21, 2024 and published in the Federal Register on August 19, 2025 (FCC 24-120), dealt with a practical issue: some providers had outsourced the signing of calls to other companies.

According to the published rule, providers with a STIR/SHAKEN obligation may use third parties to sign calls only under two conditions. The obligated provider must itself make the attestation-level decisions, and every call must be signed with that provider’s own certificate rather than the third party’s. The rule requires a written agreement describing the third party’s tasks, to be kept for two years after the arrangement ends. The obligated provider remains fully accountable.

The rule also addressed the Robocall Mitigation Database. Providers may certify full or partial STIR/SHAKEN implementation only if they hold their own token and certificate and sign calls with it; otherwise they must update their certification to say they have not implemented it, or risk referral to the FCC’s Enforcement Bureau. The FCC’s record, as cited in the rule, included an estimate that about 64% of database providers claiming implementation were not registered with the STIR/SHAKEN Policy Administrator. That is an estimate from the rulemaking record, not an audited count.

Why scam calls still get through

The FCC itself is plain about the limits. In its 2023 order it calls STIR/SHAKEN “not a silver bullet” and says expectations of stopping all illegal calls are not realistic. Several gaps explain why.

Gap Why it matters
Lawfully obtained numbers The FCC says authentication has limited effect when a number was obtained lawfully and not spoofed. A caller can be a genuine customer of a carrier and still be running a scam.
Older networks Authentication information can be lost when a call crosses non-IP (traditional) network segments. The FCC declined in 2023 to require authentication by all intermediate providers for such calls and opened an inquiry.
Not every intermediate provider must sign The 2023 rule covers only the first non-gateway intermediate provider, not each one in the chain.
Inconsistent compliance The 2025 rule’s own record points to providers claiming implementation without the registration that supports it.

In short, the system can confirm that a number was signed by a particular provider. It cannot confirm that the person speaking is honest.

Does a verified call mean a safe call?

No. A “verified” label, where one appears, generally speaks to the number, not the intentions of the caller. The FCC says authenticated caller ID information supports analytics, blocking and traceback, meaning efforts to identify the source of a call after the fact, and it describes the system as a backstop to those obligations rather than a guarantee.

This matters because a real, non-spoofed number can still belong to a fraudulent operation. The practical habits recommended in general consumer-security guidance still apply: do not act on urgency, do not share one-time codes, and verify requests by contacting the organization through a number you already know. Many of those calls and texts follow the same patterns as the messages described in how phishing works and how to spot it, and a call that leads to shared personal details can have the consequences covered in what to do after a data breach.

What to check on your own phone

Because the labels depend on carriers and devices, what you see varies. A few general points follow from how the system is built:

  1. A missing label is not proof of fraud, and a present label is not proof of legitimacy. Both are signals, not verdicts.
  2. Call-blocking and spam-filter options differ by carrier and handset. Your carrier’s website or support line can describe what it offers.
  3. Unexpected calls asking for money or codes deserve independent verification whatever the screen says. The same skepticism used to check a viral claim before sharing works for a caller’s story: pause, look for an independent confirmation, then act.

Consumers can also report unwanted or illegal calls to federal agencies such as the FCC and the Federal Trade Commission; reports feed enforcement and traceback work, though they do not guarantee that an individual call stops.

What remains unsettled

Several questions remain open. The FCC said in 2023 that it may revisit authentication across non-IP networks. Compliance with the 2025 third-party signing rule depends on dates the Wireline Competition Bureau was to announce by public notice, and how thoroughly providers follow the rules will only be visible through enforcement actions and database audits. The sources reviewed here describe the rules and the FCC’s stated rationale; they do not provide an independent measure of how many scam calls the system has actually prevented, and none was found among the sources consulted.

A seal on the number, not on the speaker

STIR/SHAKEN makes it harder to fake a phone number and easier for carriers to trace and block problem traffic. It does not certify that a caller is trustworthy, and it leaves gaps on older networks and with lawfully obtained numbers. Treating caller ID as one clue among several, rather than as proof, is consistent with how the regulator itself describes the system.

This article is general information about how phone-network rules work, not legal advice.

Sources: Federal Communications Commission; U.S. Government Publishing Office