A prospective client once asked me to analyze a phone to prove that a disputed text message was not sent from it.

I explained what an examination could do: image the phone and look for traces of the message. I also explained why that might not answer the question. Various factors could explain why a message that was once on a phone can no longer be recovered, and until the phone was imaged, nobody could say how much of its data an examiner would be able to reach. Even if an examination found nothing, I would not be able to state that the message was never on the phone, only that I had found no evidence that it was.

Then I suggested thinking about the problem differently. The phone is one witness. The carriers that carried, or did not carry, the message are others. Comparing the carriers’ records against the device is often the better evidence, and in my experience, it is the part of this kind of case most often left until it is too late.

This post explains why the phone alone rarely settles a spoofing dispute, how texts are spoofed today, what the carriers actually record, and what to ask for, and when.

Why the Phone Alone Rarely Settles It

Not every extraction reaches deleted data

How much of a phone an examiner can see depends on the type of extraction the device and the tools support. The Scientific Working Group on Digital Evidence (SWGDE) describes a logical extraction as “a process that extracts individual files or objects,” and a file system extraction as one that “extracts files from a file system and may include data marked for deletion” [1 ]. SWGDE also warns that tools “may support only a logical acquisition of data that may limit the amount of data that can be recovered” [1 ]. NIST’s mobile forensics guidelines make a similar point: a device’s memory often contains deleted data “that is not recoverable through either a logical or manual extractions” [2 ].

A full file system extraction is therefore more complete than a logical one. The deepest kind, a physical extraction, copies the storage itself, but on a modern phone it is now rare. Current phones encrypt their storage, so a physical image is unusable unless the examiner can also recover the keys: in one forensic tool vendor’s words, without them “the physical dump is theoretically just ’noise’” [3 ]. As early as 2019, a Privacy International review of extraction tools described file system acquisition as “an alternative where physical extraction of iOS devices is no longer possible” [4 ]. In my experience, a full file system extraction is now usually the most an examiner can obtain from a current phone.

Which extraction is possible depends on the age of the hardware, the operating system version, and whether the device is locked, which on a current phone also affects which encryption keys are available (see BFU vs AFU ). You will not know until the examiner tries. For more on logical and full file system extractions, see Mobile Device Extractions: A Guide for Lawyers .

Deleted messages may already be gone

Both iPhones and Android phones store text messages in SQLite databases. On an iPhone, the sms.db file holds SMS, MMS and iMessage content together [5 ]; on Android, the equivalent is mmssms.db [6 ].

The SQLite documentation explains why deleted messages can sometimes be recovered: “When content is deleted from an SQLite database, the content is not usually erased but rather the space used to hold the content is marked as being available for reuse. This can allow deleted content to be recovered by a hacker or by forensic analysis” [7 ]. The same page explains why they sometimes cannot: running the VACUUM command rebuilds the file and cleans it “of all traces of deleted content” [7 ]. Belkasoft, a forensic tool vendor, describes the same mechanics from the examiner’s side: deleted records can survive in free pages and unallocated space inside the database, and VACUUM “will rebuild the database and truncate it” [8 ].

There is a third place to look. Databases running in write-ahead log (WAL) mode do not change the main file right away: “The original content is preserved in the database file and the changes are appended into a separate WAL file” [9 ]. Older versions of a record can survive in the WAL until a checkpoint folds it back into the database. Whether any of this survives on a particular phone is an empirical question that only the examination answers.

Absence of evidence is not evidence of absence

This is the limit attorneys most need to hear before the examination, not after. If the message is not found, that is a finding about what the examination could see. It is not proof the message never existed. The UK Forensic Science Regulator, in a 2014 consultation draft on method validation, put the general principle this way in the context of cell-site analysis: without validation of how often a method misses things, “failure to detect a cell at a location may not, in itself, indicate that the phone could not have been there” [10 ]. The same caution applies to a missing text message.

The phone is still worth examining. A fake-text app, for example, can leave its own fingerprints: one examiner found that a forensic image showed a disputed message was created by an app called “FakeTextMessage” [11 ]. But a clean result on the phone answers a narrower question than the one the case is asking.

How a Text Gets Spoofed Today

Spoofing a text means making a message display a sender it did not come from. The wireless industry’s own guidelines define message number spoofing as causing a message “to display an originating number for the message that is not assigned to the Message Sender,” or originating it through a provider other than the one that receives replies [12 ]. It is not a theoretical problem. Federal law makes it unlawful to transmit misleading caller identification “in connection with any voice service or text messaging service” with intent to defraud, cause harm or wrongfully obtain anything of value [13 ]. The same statute defines a text message to include SMS and MMS [13 ]. The FCC’s implementing order expressly reaches spoofing through “alternative voice and text messaging services” [14 ].

The common routes are:

  • Messaging services that are not the number’s own carrier: the second half of the industry definition above describes this route, in which the message is originated through a provider other than the one that holds the displayed number [12 ]. It follows that the recipient sees a familiar number, while the purported sender’s carrier would have no outbound record of the message from that line.
  • Email-to-text gateways: the FCC has recorded that “email-to-text gateways enable anyone to send a text message to a mobile subscriber in relative anonymity,” and called email-to-text “a major source of illegal texts” [15 ].
  • Fake-text apps and fabricated screenshots: an examiner has found a disputed message that a fake-text app created on the phone itself [11 ]. Apps that fabricate a message locally, and fabricated screenshots, never touch a carrier at all. It follows that such a message exists only on one device, or only in an image.

Three things are often confused with spoofing and are not the same:

  • SIM swap and port-out fraud: here a bad actor takes over the victim’s actual number, which “gives the bad actor control over the customer’s phone number” [16 ]. Because messages sent this way travel on the real line, it follows that carrier records would show them as sent from that number, even though someone else sent them.
  • Messages composed on a paired computer: with Apple’s Text Message Forwarding, the SMS messages an iPhone sends and receives also appear on a paired Mac or iPad, “so you can continue the conversation from the device you want” [17 ]. Google Messages for web sends SMS “using a connection from your computer to your phone” [18 ]. The phone relays the message, so the carrier sees the phone, even though the keyboard was somewhere else.
  • iMessage and RCS: these are not SMS. Apple’s iMessage is sent “over Wi-Fi or cellular-data networks” [19 ], and Google’s RCS chats are sent “over Wi-Fi and mobile data” [20 ]. A blue bubble or an RCS chat will generally not appear in a carrier’s text message records at all, which matters a great deal when you read those records.

What the Carriers Record

Depending on the carrier, its records for each text message may include the date, the time and the sending and receiving phone numbers, along with location [11 ]. These detail records are metadata, not the words of the message. (Some carriers have briefly kept content too; whether it can be reached by subpoena is a legal question discussed below.) For calls, these are the Call Detail Records most attorneys have seen. Text messages have a parallel set of records, and the FCC has observed that wireless providers “already know the transmitting provider (e.g., wireless provider or messaging solutions aggregator) and user identifier” for the texts they deliver [21 ].

That observation suggests where to look in a spoofing case, though whether a carrier retains that information, and in what form, is something to ask for, not assume. Two carriers may hold records of the same disputed message, and they answer different questions:

  • The recipient’s carrier records the message as it arrived, including the number it displayed. If the sender spoofed the number, the recipient’s carrier record would ordinarily show the spoofed number. It cannot by itself tell you who sent it, although the upstream provider it arrived from can be a valuable lead.
  • The purported sender’s carrier records an outbound message from that line only if one actually left it. If the message was spoofed through a gateway, the sender’s carrier would have no outbound record of it from that line. If the message really was sent from that line as SMS, we would expect one, provided the carrier still held records of that type for that date.

Three cautions keep this comparison honest. First, it covers SMS and MMS only. If the message was an iMessage or an RCS chat, neither carrier’s text records will ordinarily show it, and its absence there tells you little. Establish the channel first, from the recipient’s device or a forensic image of it. Second, normalize the time zones. Confirm which time zone each set of records uses, because it may not match the one on the phone’s screen, and a mismatch of a few hours can make a real message look missing or a missing one look real. Third, confirm the records exist: that the carrier still retained that record type for that date, that the production covers the right line and window, and that the subpoena went to the network that carried the traffic, not only a reseller.

Caller ID authentication covers calls, not texts

Attorneys sometimes ask whether STIR/SHAKEN (Secure Telephone Identity Revisited and Signature-based Handling of Asserted Information Using toKENs), the caller ID authentication framework, answers the question. For calls, it can help: the FCC required voice providers to implement it in their IP networks by June 30, 2021, and an originating provider attests whether it can confirm the caller and the number (A, B or C attestation) [22 ]. As of the FCC’s 2023 order, it does not apply to text messages. The FCC itself noted that “STIR/SHAKEN only works on the Session Internet [sic] Protocol (SIP),” and declined to adopt a caller ID authentication requirement for texts [21 ]. As of that order, there is no authentication flag to check for a disputed text; the carrier records and the devices are where the answer lies.

What to Ask For, and When

Preserve first

Carrier retention is limited and varies by carrier and by record type. A 2010 Department of Justice chart, made public through an ACLU records request [23 ], showed text message detail kept anywhere from one year to seven years depending on the carrier, and text message content kept for a few days, up to 90 days at one carrier, or not at all [24 ]. Those figures are historical, and current policies may differ, so confirm current retention from the carrier before relying on any period. The lesson has not changed: the clock is running from the day of the message.

Send preservation letters to both carriers and to the opposing party as soon as the dispute surfaces. The statutory preservation demand under 18 U.S.C. § 2703(f) is available only to a “governmental entity” [25 ], so a private litigant’s letter generally depends on the carrier’s cooperation and on the opposing party’s duty not to destroy evidence. It is still the cheapest step in the case.

What a civil subpoena can and cannot reach

The Act prohibits a provider from divulging “the contents of a communication while in electronic storage” [26 ]. At least one federal court has quashed civil subpoenas that sought private message content from online providers on that basis [27 ]. Non-content records are different: the same statute permits a provider to disclose non-content records pertaining to a subscriber “to any person other than a governmental entity” [26 ]. It permits; it does not compel.

Content is reachable through the parties instead: the statute permits disclosure “with the lawful consent of the originator or an addressee or intended recipient” (§ 2702(b)(3)) [26 ], and a party’s own device and account data are ordinary discovery. If your client is the recipient, their own carrier records may need no subpoena at all. The records a spoofing comparison needs are non-content records.

A practical request list, adjusted to the facts:

  1. From the purported sender’s carrier: SMS and MMS detail records (and call detail records, if calls are also in dispute) for the line, for a window around the disputed message, with the time zone stated.
  2. From the recipient’s carrier: inbound SMS and MMS detail records for the same window, including the originating number as delivered and any information identifying the upstream provider or aggregator that handed off the message.
  3. From the opposing party, through discovery: the device, or a forensic image of it, for examination, and any iCloud or Google account data that would show the message’s channel.
  4. From any upstream provider the recipient’s carrier identifies: its records for the message, which can lead to the account that actually sent it.

In criminal cases the routes differ, and the defense is in no better position than a civil litigant to reach content. The government obtains carrier records by subpoena, court order or warrant under 18 U.S.C. § 2703 and can send a § 2703(f) preservation request. The defense cannot, so ask the prosecutor early whether records were obtained or preserved, and consider asking the court for a preservation order. The defense can seek non-content records by subpoena under the jurisdiction’s criminal subpoena rule, which often requires court approval and a showing of relevance and specificity. The underlying point is the same: the comparison only works if both sides of the message are in the record.

Reading the Comparison

When the records come back, the patterns are usually readable:

  • The recipient’s carrier shows an inbound SMS from the number, and the sender’s carrier shows a matching outbound SMS at the same moment. The message went over that line. It does not yet tell you whose hand was on the phone, whether the phone was relaying a message typed on a paired computer, or whether the line itself had been taken over by a SIM swap or port-out.
  • The recipient’s carrier shows an inbound SMS displaying the number, and the sender’s carrier shows nothing. That is consistent with spoofing, and the upstream provider becomes the next lead. It is an indicator to be tested, not proof: confirm the channel was SMS and the time zones were aligned, and confirm the records exist (that the carrier still retained that record type for that date, that the production covers the right line and window, and that the subpoena went to the network that carried the traffic, not only a reseller) before relying on it.
  • Neither carrier shows the message. Consider iMessage, RCS, an app, or a fabricated screenshot, and look to the devices.

Each pattern narrows the question. None of them, on its own, answers who sent the message. That is a question about people, and it is answered with everything else in the case, not with one record.

I have supported several cases in which SMS spoofing was used to make a message appear to come from someone who had not sent it, so that the recipient would retaliate against that person. In those cases, the most useful evidence came from putting the device and the carrier records side by side, not from either one alone.

If your case turns on whether a text was really sent, Lucid Truth Technologies can examine the devices, help counsel identify which carrier records to request and read the records against each other.

References

Sources, statutes and URLs in this post were verified on 2026-11-18.

[1] Scientific Working Group on Digital Evidence, “SWGDE Best Practices for Mobile Device Evidence Collection & Preservation, Handling, and Acquisition,” v1.2, Sep. 17, 2020. [Online]. Available: https://www.swgde.org/wp-content/uploads/2023/11/2020-09-17-SWGDE-Best-Practices-for-Mobile-Device-Evidence-Collection-_-Preservation-Handling-and-Acquisition_v1.2.pdf

[2] R. Ayers, S. Brothers and W. Jansen, Guidelines on Mobile Device Forensics, NIST SP 800-101 Rev. 1, National Institute of Standards and Technology, May 2014. [Online]. Available: https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-101r1.pdf

[3] V. Katalov, “Perfect Acquisition: The True Physical Acquisition,” ElcomSoft Blog, Feb. 16, 2026. [Online]. Available: https://blog.elcomsoft.com/2026/02/perfect-acquisition-the-true-physical-acquisition/

[4] Privacy International, “A Technical Look at Phone Extraction,” Oct. 2019. [Online]. Available: https://privacyinternational.org/sites/default/files/2019-10/A%20technical%20look%20at%20Phone%20Extraction%20FINAL.pdf

[5] Magnet Forensics, “The Meaning of Messages,” Magnet Forensics Blog, Mar. 9, 2023. [Online]. Available: https://www.magnetforensics.com/blog/the-meaning-of-messages/

[6] J. McQuaid, “Android Messaging Forensics: SMS/MMS and Beyond,” Magnet Forensics Blog, Sep. 11, 2018. [Online]. Available: https://www.magnetforensics.com/blog/android-messaging-forensics-sms-mms-and-beyond/

[7] SQLite, “VACUUM,” SQLite Documentation. [Online]. Available: https://www.sqlite.org/lang_vacuum.html

[8] Belkasoft, “SQLite Forensics with Belkasoft X.” [Online]. Available: https://belkasoft.com/sqlite-forensics-with-belkasoft-x

[9] SQLite, “Write-Ahead Logging,” SQLite Documentation. [Online]. Available: https://www.sqlite.org/wal.html

[10] UK Forensic Science Regulator, “Draft Guidance: Digital Forensics Method Validation,” consultation draft, Aug. 2014, sec. 13.2. [Online]. Available: https://assets.publishing.service.gov.uk/media/5a7df712e5274a2e87daeaab/2014.08.28_FSR_Digital_validation_Draft.pdf

[11] A. Merlino, “Assuring Your Text Messaging Evidence Is Not Spoofed,” Vestige Digital Investigations, Jun. 19, 2019. [Online]. Available: https://vestigeltd.com/resources/articles/assuring-your-text-messaging-evidence-is-not-spoofed

[12] CTIA, “Messaging Principles and Best Practices,” May 2023, sec. 6.2. [Online]. Available: https://api.ctia.org/wp-content/uploads/2023/05/230523-CTIA-Messaging-Principles-and-Best-Practices-FINAL.pdf

[13] 47 U.S.C. § 227(e). [Online]. Available: https://www.law.cornell.edu/uscode/text/47/227

[14] Federal Communications Commission, Implementing Section 503 of RAY BAUM’S Act; Rules and Regulations Implementing the Truth in Caller ID Act of 2009, Second Report and Order, FCC 19-73, Aug. 5, 2019. [Online]. Available: https://docs.fcc.gov/public/attachments/FCC-19-73A1.txt

[15] Federal Communications Commission, Targeting and Eliminating Unlawful Text Messages, Second Report and Order, FCC 23-107, Dec. 18, 2023. [Online]. Available: https://docs.fcc.gov/public/attachments/FCC-23-107A1.pdf

[16] Federal Communications Commission, Protecting Consumers from SIM-Swap and Port-Out Fraud, Report and Order, FCC 23-95, Nov. 16, 2023. [Online]. Available: https://docs.fcc.gov/public/attachments/FCC-23-95A1.pdf

[17] Apple, “Forward text messages from your iPhone to other devices,” Apple Support. [Online]. Available: https://support.apple.com/en-us/102545

[18] Google, “Check your messages on your computer or Android tablet,” Google Messages Help. [Online]. Available: https://support.google.com/messages/answer/7611075?hl=en

[19] Apple, “What is the difference between iMessage, RCS, and SMS/MMS?,” Apple Support. [Online]. Available: https://support.apple.com/en-us/104972

[20] Google, “RCS chats by Google FAQ,” Google Messages Help. [Online]. Available: https://support.google.com/messages/answer/9487020?hl=en

[21] Federal Communications Commission, Targeting and Eliminating Unlawful Text Messages, Report and Order and Further Notice of Proposed Rulemaking, FCC 23-21, Mar. 17, 2023. [Online]. Available: https://docs.fcc.gov/public/attachments/FCC-23-21A1.pdf

[22] Federal Communications Commission, Call Authentication Trust Anchor, Report and Order and Further Notice of Proposed Rulemaking, FCC 20-42, Mar. 31, 2020. [Online]. Available: https://docs.fcc.gov/public/attachments/FCC-20-42A1.pdf

[23] ACLU of North Carolina, “How Long Is Your Cell Phone Company Hanging On To Your Data?,” Sep. 28, 2011. [Online]. Available: https://www.acluofnorthcarolina.org/news/how-long-your-cell-phone-company-hanging-your-data/

[24] U.S. Department of Justice, Computer Crime and Intellectual Property Section, “Retention Periods of Major Cellular Service Providers,” Aug. 2010. [Online]. Available: https://www.enriquedans.com/wp-content/uploads/2011/10/dataretentioncellphoneco.pdf

[25] 18 U.S.C. § 2703(f). [Online]. Available: https://www.law.cornell.edu/uscode/text/18/2703

[26] 18 U.S.C. § 2702. [Online]. Available: https://www.law.cornell.edu/uscode/text/18/2702

[27] Crispin v. Christian Audigier, Inc., 717 F. Supp. 2d 965 (C.D. Cal. 2010). [Online]. Available: https://www.jenner.com/a/web/iKifMkGDUonDNqjpSfwF8o/4HRMZQ/Crispin_v._Audigier.pdf