US 12554825

I'll search for this specific patent number across USPTO and litigation sources.

Let me dig deeper for the specific patent's claims and any 2026 CAFC activity.

Let me try to locate the published application and any 2026 Federal Circuit activity.

Sourcing note

I ran targeted searches for the literal string 12554825 / "12,554,825" against general web indexes (which surface USPTO/Google Patents, Justia, Unified Patents, PTAB dockets, Darts-ip, and CAFC filings). Results returned:

• No direct hit on the number US 12,554,825 in any index I could reach.
• Multiple hits on sibling patents in the same family (same title, same inventor, same assignee), which corroborate the bibliographic data.
• One 2026 Federal Circuit appellate filing involving Proxense, but on a different patent (see below).

I am therefore relying primarily on the authoritative full text supplied in the prompt (Google Patents rendering of US12554825B2, fetched 2026‑09‑30) for bibliographic facts, and I flag below where I lack authoritative data.

⚠️ Date discrepancy: your instruction says "Current Date: April 26, 2026," but my operating environment and the source fetch timestamp are 2026‑09‑30. I have not reconciled these; treat the bibliographic dates below as the operative ones.

1. Verified bibliographic summary (US 12,554,825 B2)

Corroborating family members (same title/inventor/assignee, useful for tracing):

• US 10,469,456 B1 — issued 2019‑11‑05, app 14/973,565, 18 claims
• US 11,086,979 B1 — issued 2021‑08‑10, app 16/672,214, 20 claims
• US 12,554,825 B2 — app 17/394,837 (the present patent)

There is also a US 9,251,332 B2 in the same family tree (cited in the priority-applications listing), consistent with a long continuation chain from the 2007 disclosure.

2. Abstract (verbatim from the authoritative text)

A security system comprises a personal digital key (PDK), a reader and a computing device. The PDK is a portable, personal transceiver that includes a controller and one or more passwords or codes. The computing device includes a detection engine, vault storage and a set up module. The detection engine detect events relating to the access of any files and third‑party systems by the computing device and receives information from the reader as to whether the PDK is present/linked. The detection engine controls whether a user is able to access any of the functionality provided by the computing device based upon whether the PDK is in communication with the reader or not. The present invention also includes a number of methods such as a method for initializing the security system, a method for setting up a computing device, and a method for controlling access to computing resources.

3. Independent claims — ⚠️ I do NOT have authoritative claim text

Important limitation: the patent text provided to me was truncated before the claims section (it ends mid‑sentence in the FIG. 11D description of step 1150). Neither Google Patents' rendering as supplied, nor any search result, gave me the actual claim set of US 12,554,825. I therefore cannot quote or faithfully paraphrase its independent claims, and I will not fabricate them.

What I can state with confidence is the disclosed subject matter that the independent claims are drawn to, from the specification's own summary and the FIG. 11A–11E flowchart:

Likely independent-claim subject matter (based on the disclosure, not the claims):

1. System claim — a combination of (a) a personal digital key (PDK) that is a portable wireless transceiver storing passwords/codes, (b) a reader that wirelessly links to the PDK and provides a wired output to, and (c) a computing device having a detection engine, vault storage, and a set up module; wherein the detection engine detects events (storage‑device/directory/file access, sign‑on screen display, time‑based triggers) and enables or disables access based on whether the PDK is in communication with the reader.

2. Method claim for controlling access to computing resources — monitoring for an event; determining event type; retrieving a corresponding item set‑up record from vault storage; validating the PDK via the reader (e.g., PDK within range); and allowing or precluding access based on validation, and for sign‑on screens, auto‑filling and submitting credentials from the set‑up record.

3. Method claim for initializing / setting up the system — connecting the reader to the computing device, downloading and running the reader's setup module on the computing device, creating vault storage, loading the detection engine, linking a PDK (first PDK becomes master PDK with authority to proxy rights to other PDKs), creating and storing item set‑up records, and storing sign‑on login information on the PDK.

Distinguishing features the disclosure emphasizes (likely reflected as limitations): the ~5‑meter (scalable) RF range allowing the PDK to stay in a pocket/purse; hierarchical set‑up records (file > directory > device); time‑dependent exit rules with a duration timer triggering auto‑lock/logout/screensaver; a backup password for PDK‑absent access; proxy/grouped PDKs; and optional biometric reconfirmation.

I would need the issued claims (e.g., from USPTO PatentCenter/Patent Public Search or the printed patent PDF) to give you an accurate per‑independent‑claim breakdown and claim count.

4. Litigation / CAFC 2026 search results

• No CAFC 2026 docket specifically naming US 12,554,825 was found. Because the patent issued 2026‑02‑17, an appeal involving this patent would be extraordinarily early as of your stated April 2026 date — this is a plausible explanation for the null result, though I cannot rule out a recently filed/not‑yet‑indexed docket.
• Relevant Proxense appellate activity found (different patent): a Federal Circuit opening brief dated 2026‑04‑03 in Proxense v. Apple / Microsoft (hosted on ip fray: ipfray.com/wp-content/uploads/2026/04/26-04-03-Proxense-v.-Apple-Microsoft-appellate-opening-brief-1.pdf). This concerns US 8,646,042 B1 (and related '730/'954/'905/'989/'042/'188 patents), following the IPR2024‑00573 Final Written Decision (Microsoft Corp. and Apple Inc. v. Proxense, LLC) holding claims 1, 5, 6, 8–11, 13, 14 of the '042 patent unpatentable under § 103. That is a different patent family branch (PDK/RDC hybrid‑device biometric patents), not the '825 patent.
• District court activity: Proxense, LLC v. Apple, Inc., No. 6:24‑cv‑00143 (W.D. Tex.) and Proxense, LLC v. Microsoft Corp., No. 6:23‑cv‑00319 (W.D. Tex.) / Proxense v. Google, No. 6:23‑cv‑00320 — again asserted patents in the biometric/PDK family, not confirmed to include the '825 patent.

5. Explicit uncertainty statement

• High confidence: title, application number 17/394,837, filing date 2021‑08‑05, inventor John J. Giobbi, assignee Proxense LLC, priority 2007‑12‑19, grant date 2026‑02‑17, abstract text, and the continuation chain. These come from the authoritative patent text in hand.
• Medium confidence: "Active" status and the 2028‑12‑19 anticipated expiration (these are Google Patents' stated assumptions, not legal conclusions; maintenance fees and PTA could change the picture).
• Low confidence / unknown: the actual claims of US 12,554,825 (count, numbering, exact limitations) — the text supplied was truncated before the claims, and no search result surfaced them. Any litigant, examiner, or clearance opinion should pull the issued claims directly from USPTO PatentCenter / Patent Public Search (patft) rather than rely on the inference section above.
• Unknown: whether US 12,554,825 has been asserted in any litigation or is subject to any IPR/PGR as of the stated date. The Darts‑ip flag applies to the family, not confirmed to this specific number.

If you can supply the claims section (or authorize me to work from the sibling US 11,086,979 B1 / US 10,469,456 B1 claims as a proxy), I can give you the precise independent‑claim breakdown you asked for.

I'll search for litigation involving US patent 12554825 across patent litigation sources.

Let me search more specifically for the assignee Proxense's litigation and check the litigation databases.

I have substantial Proxense litigation data. Now let me verify specifically whether US 12,554,825 itself has been asserted.

Summary: Litigation Involving U.S. Patent No. 12,554,825

Bottom line: I found no litigation specifically asserting U.S. Patent No. 12,554,825.

What I did and did not find

I searched patent-litigation sources (Justia Dockets, Docket Alarm, RPX Insight, Unified Patents, CourtListener, PTAB/PTActs, Law360, ip fray) for the specific number 12,554,825 and for the assignee (Proxense LLC). No complaint, docket entry, PTAB petition, or news item identified any case naming U.S. 12,554,825 as a patent-in-suit.

This is consistent with the patent's chronology as reflected in the text you provided:

• Patent: U.S. 12,554,825 B2 — "Security system and method for controlling access to computing resources"
• Assignee: Proxense LLC (inventor John J. Giobbi)
• Application: 17/394,837, filed 2021-08-05
• Publication of application: US20210365537A1, 2021-11-25
• Grant date: 2026-02-17
• Priority date: 2007-12-19 (continuation chain: 16/672,214 → 14/973,565 → 12/340,501 → provisional 61/015,110)

Because the patent issued only recently (February 17, 2026), any suit asserting it would be very new, and no such case surfaced in the sources I could reach.

Important caveat / verification note

• The Google Patents page you supplied carries a "Family has litigation" flag (linking to Darts-ip family 40790320). That flag attaches to the patent family, not necessarily to this particular patent. Proxense's family includes dozens of related patents, many of which have been litigated — so a "family has litigation" flag is not evidence that 12,554,825 itself has been asserted. I could not independently confirm, from the sources retrieved, any case in which 12,554,825 is a patent-in-suit.
• I was unable to complete additional targeted searches (search limits were reached) of the Unified Patents litigation case list, PACER, and the CAFC docket specifically keyed to "12554825." I therefore state this with moderate confidence, not certainty.

Proxense family litigation (context — NOT confirmed to involve the '825 patent)

For background on the assignee's enforcement activity, the following Proxense cases were identified. None of these were shown to assert U.S. 12,554,825; the patents asserted in them (e.g., 8,352,730; 8,886,954; 9,298,905; 8,646,042; 9,679,289; 10,073,960; 10,698,989; 9,049,188; 9,235,700; 8,340,672) are different from the '825 patent.

Additionally, multiple PTAB/IPR proceedings were filed by Apple, Google, Microsoft, and Samsung against various Proxense patents (e.g., IPR2021-01439; IPR2024-00233; IPR2024-00846; IPR2024-01327; IPR2024-01334; IPR2024-01485; IPR2024-01399), and Proxense filed an appellate matter (In re Google LLC, No. 24-117). Again, none of these were shown to involve U.S. 12,554,825.

Recommendation

To confirm definitively (since I could not complete all targeted queries):

1. Search PACER and Unified Patents' litigation case list by the patent number "12554825."
2. Check Docket Alarm / RPX Insight filtering by patent number rather than by party.
3. Confirm the family flag via Darts-ip family 40790320 to see whether the specific '825 patent (as opposed to sibling patents) is tagged.

If those searches also return nothing, the accurate answer is that no litigation is known asserting U.S. 12,554,825 as of the current date — while noting the assignee (Proxense) is a highly active litigant on sibling patents in the same family.

I'll start by verifying the structured block against what's publicly indexed on the web, since the ODP ingest can lag recently-filed petitions.

The web results show a large Proxense IPR campaign, but all hits so far are on different patents in the family ('289, '960, '042, '188, '700, '954, '730). Let me target US 12,554,825 specifically and its immediate parents.

Important structural finding emerging. Let me verify the family tree and check for proceedings against the sibling patents in this exact chain ('332, '456, '979).

Proceedings overview

Zero AIA trial proceedings are on file for US 12,554,825. The structured PTAB block (USPTO ODP) returns no IPR, PGR, or CBM proceeding naming this patent, and my web searches surfaced nothing that contradicts that — every Proxense PTAB hit was directed at a different patent number. The breakdown is therefore 0 active / 0 invalidated / 0 sustained / 0 settled / 0 denied.

That "zero" is not a merits signal — it is a statutory-timing artifact. US 12,554,825 granted on 2026-02-17 (filed 2021-08-05 as a continuation; priority 2007-12-19). Under 35 U.S.C. § 311(c), an IPR petition may not be filed until the later of (i) nine months after grant or (ii) termination of an instituted PGR. Nine months from grant is 2026-11-17. As of today (2026-09-30), the IPR window on the '825 has not yet opened. A PGR could have been filed in the 2026-02-17 → 2026-11-17 window; none appears on file.

Bottom line for a defendant today: there is no IPR history to lean on and no estoppel to exploit — but also no adverse PTAB ruling to worry about. You are, however, standing on the front edge of a patent that is almost certainly IPR-eligible in ~7 weeks, held by an owner (Proxense) with an active, well-documented PTAB war room and a serial litigator's posture.

Proceedings on US 12,554,825

None. There is no proceeding to itemize. Per the operating constraints, I will not manufacture proceeding numbers to fill this section.

Adjacent PTAB activity that matters (NOT proceedings on the '825)

Flagging these because they bear on how the '825 will be attacked — but stating plainly: none of these is a proceeding against US 12,554,825, and none creates § 315(e)(2) estoppel against a would-be '825 petitioner.

Same-family siblings (the "ProxAccess" chain)

The '825 sits on a straight continuation chain, all sharing the 2007-12-19 priority date and the same specification:

Defensive-aggregator signal: Unified Patents' PATROLL program ran a prior-art bounty (US$2,000) specifically on claim 1 of US 9,251,332 — the grandparent of the '825 — describing it as "a security system and method for controlling access to computing resources, featuring the use of personal digital keys (PDKs) and readers," and noting it "has been asserted against Hyundai and LG." The contest closed 2025-08-12 and a winner was announced 2025-10-23 (PATROLL contest; Unified Patents award announcement). That is a live, funded invalidity effort aimed at this exact specification. The collected art is public and directly reusable against the '825's claims, which share the same disclosure. Verification caveat: I confirmed the contests, not whether any resulting IPR was actually filed against the '332, '456, or '979 — my family-specific search was truncated before I could exhaust it, so treat "no IPR on the siblings" as unverified rather than established.

Different Proxense families (context on the owner's PTAB exposure)

The bulk of Proxense's PTAB docket involves other patent families (hybrid device / biometric PDK), not the '825's chain:

• US 8,886,954 — IPR2024-00233 and IPR2024-00846 (Microsoft); FWD found claims 1, 2, 4, 5, 7-13, 15, 16, 18-24, and 26-29 unpatentable under pre-AIA § 103; on appeal to the Federal Circuit (Proxense's opening brief filed 2026-04-03, raising an APA/§ 706 challenge to the Board's treatment of parallel ex parte reexamination 90/015,052). See IPR2024-00233 institution decision.
• US 8,352,730 — IPR2024-00775 (Microsoft), terminated after Patent Owner request for adverse judgment; also in ex parte reexam 90/015,052.
• US 9,679,289 — IPR2024-00407 (Microsoft, instituted 2024-07-22), IPR2024-00783 (Google, instituted 2024-11-18), IPR2025-00075/00074 (Apple, joinder), IPR2024-01319 (Google, institution denied as copycat).
• US 10,073,960, 8,646,042, 9,542,595, 9,298,905, 9,049,188, 8,886,700 — additional IPRs/reexams from Microsoft, Google, Apple, Samsung.

Pattern signals: (1) the same petitioners (Microsoft, Google, Apple) file serial and copycat petitions across Proxense's portfolio, using joinder to consolidate; (2) Proxense litigates hard and appeals adverse FWDs to the Federal Circuit rather than settling; (3) a defensive aggregator (Unified) is actively funding prior art on the '332. All three point the same direction: an IPR against the '825 on or shortly after 2026-11-17 is a realistic, arguably likely event.

Strategic summary

Claim status on the '825: entirely UNTESTED. No claim of US 12,554,825 has been canceled, confirmed, or even challenged in an AIA trial. There is no claim-level narrowing to rely on and no PTAB construction to borrow. Unlike its cousin US 8,886,954 — where the Board canceled claims 1, 2, 4, 5, 7-13, 15, 16, 18-24, and 26-29 — the '825 is a clean slate. Do not let anyone tell you the '825 has been "invalidated"; it has not been touched.

Estoppel: nothing to work with. § 315(e)(2) estoppel runs only against petitioners (and privies) who were in an IPR on this patent. With zero '825 petitions, every ground — § 102 anticipation, § 103 obviousness, § 112 written description/enablement/indefiniteness, and (if eligible within the PGR window) § 101 — remains fully available in the district court. The flip side: nothing stops you from filing the first petition, and nothing binds you to anyone else's prior art strategy.

Two constraints shape the playbook, and both cut in your favor as a defendant. First, § 315(b): the one-year bar runs from service of a complaint alleging infringement of the '825, so track your service date. The clock to file an IPR is the earlier of that anniversary and the practical case schedule. Second, short remaining life: the '825 carries an anticipated expiration of 2028-12-19 and the family is characterized by terminal disclaimers tying later continuations to the earliest link in the chain. That leaves roughly 2.2 years of term from today. A petition filed in late 2026 would very likely reach an FWD (statutory 1-year trial deadline from institution) near or after expiry — a strong practical argument for a Sotera-style stipulation to neutralize Fintiv and for leveraging the short runway in damages/royalty negotiations.

Recommended next steps

1. Calendar 2026-11-17 — the earliest date any IPR can be filed on the '825, and simultaneously the PGR deadline. If a PGR is filed in the next ~7 weeks, the IPR clock can slide further (IPRs then open only at PGR termination). Monitor PTAB E2E and the ODP for the first petition; a first petition will spawn a § 315(a)/(b)/(e) estoppel web that materially changes your risk calculus.

2. Pull the Unified Patents winning art on US 9,251,332 now. Because the '825 shares the '332's specification and 2007-12-19 priority, the funded prior-art charts are a free head start on an invalidity position — subject to confirming the '825's actual claim scope, which I have not verified.

3. Run your own § 315(b) service-date audit and decide whether you want to be the first petitioner. Being first avoids General Plastic follow-on estoppel and Fintiv discretion, and lets you shape the Board's claim constructions — but it also means bearing the full cost without the benefit of an earlier FWD.

4. Preserve § 112 and § 101 grounds. These are outside the IPR lane (IPR is limited to §§ 102/103 on patents and printed publications under § 311(b)), so they survive any adverse IPR outcome. Indefiniteness and § 101 challenges on a 2007-priority software/security claim should be developed in parallel.

5. Do not cite any "PTAB ruling" on the '825 in correspondence — there is none. If opposing counsel or a demand letter implies prior PTAB validation of the '825's claims, that is false; the patent simply has not been tested.

Verification and limitations. The proceeding count of zero is grounded in the structured ODP block and corroborated by search. Key dates (grant 2026-02-17; App. 17/394,837 filed 2021-08-05; priority 2007-12-19; anticipated expiration 2028-12-19) are from the Google Patents record. The § 311(c) nine-month analysis is statutory. I could not complete verification of whether IPRs exist against the immediate parents US 9,251,332, US 10,469,456, or US 11,086,979, nor confirm judge panels for any of the adjacent proceedings — treat those items as open. Specific facts about the '954 FWD and the Federal Circuit appeal are drawn from third-party reproductions of the Board's decisions and Proxense's appellate brief, not from the PTAB's own docket; verify against PTAB E2E and the Federal Circuit docket before relying on them.

I'll research the assignment chain and litigation history for US 12554825. Let me run several searches.

Let me dig into the assignment records and counsel of record.

Ownership & NPE Analysis — US 12,554,825 B2

"Security system and method for controlling access to computing resources" · Appl. 17/394,837 · filed 2021-08-05 · granted 2026-02-17 · anticipated expiration 2028-12-19

Data-access note (read first): My search tooling could not open USPTO Assignment Center (assignmentcenter.uspto.gov) or assignment.uspto.gov — every query against those endpoints returned empty. I therefore could not read reel/frame numbers, conveyance codes, or assignment correspondents for this patent. What follows is built on the Google Patents legal‑events record for this number (authoritative text supplied) plus court, PTAB, and NPE‑directory records. I have flagged every point where reel/frame citations are missing rather than inventing them.

Inventors

• The patent is a single-inventor continuation. No co-inventors are named on US 12,554,825 or on the parent '456 patent in the same family.
• Giobbi was not a departing inventor: he remained CEO of the assignee through the entire 2007→2026 prosecution history. Proxense's complaint describes him as founder/CEO and notes prior roles at WMS Gaming (SVP) and Prelude Technology Corp. (source: Proxense v. Microsoft, W.D. Tex. 6:23‑cv‑00319, Dkt. 1, ¶36).
• Unusual pattern actually present: not inventor churn, but continuation churn with a single inventor and a single prosecuting firm. Application 17/394,837 is the 4th link in a chain (12/340,501 → 14/973,565 → 16/672,214 → 17/394,837), and the family keeps spawning siblings (US 10,469,456 B1; US 2025/0165580 A1, published 2025‑05‑22, same title). This is a licensor-pattern refiling strategy, not a product-development pattern.

Original assignee

• Proxense, LLC (Bend, OR) — both the original and the current assignee of record on Google Patents; the legal‑events list for US 12,554,825 shows only filing/prosecution/priority events, no assignment events at all (source: https://patents.google.com/patent/US12554825/en).
• History: incorporated in 2001 as Margent Development LLC (later renamed Proxense), founded by John Giobbi. Per the complaint it developed "ProxPay" mobile-payment and PDK (Personal Digital Key) proximity-authentication technology from ~2004–2012, employing 30+ engineers and spending millions on R&D.
• Did it ship a product embodying the claims? It claims to have commercialized PDK/ProxPay technology and demoed a "wireless wallet" (The Bulletin, 2008; DARKReading, 2009), but there is no evidence of any product in commerce today, and Stanford's NPE Litigation Database classifies the entity as an NPE of type "3 Failed startup" (https://npe.law.stanford.edu/case/[286643](/patent/286643)).
• Pivotal event: on 2016‑07‑25, Proxense's outside IP counsel wrote to Apple (Tim Cook) offering to sell the entire patent portfolio in a single transaction, stating: "Proxense has never engaged in any patent enforcement litigation. Additionally, there are no current licenses granted under the Proxense patents. Therefore, the ultimate purchaser will be able to use the Proxense patents offensively against others." (letter attached as Dkt. 1‑39 in Proxense v. Apple, 6:24‑cv‑00143; https://www.docketalarm.com/cases/Texas_Western_District_Court/6--24-cv-00143/Proxense_LLC_v._Apple_Inc/docs/1/39.pdf)
• Current status: operating as a patent-holding / enforcement entity — not dissolved, not in bankruptcy, still prosecuting new continuations (this patent issued Feb. 2026). Serial litigant since 2021 (Samsung, Google, LG, Hyundai, Microsoft, Apple, Intel).

Assignment timeline

Finding: no post‑issuance or third‑party assignment is recorded for US 12,554,825. Google Patents' legal-events list for this number contains no assignment entries, and the "original assignee" and "current assignee" fields are the same entity — Proxense, LLC. The rights flow originally from inventor Giobbi to Proxense (the standard founder→company assignment that would have been recorded against the 2008 ancestor application, 12/340,501), and the chain terminates there.

Because I could not open Assignment Center, I cannot give reel/frame numbers, execution dates, or the correspondent of record. Per your instructions, I am stating that plainly — but I am continuing past this section because the NPE determination below rests on independent, verifiable evidence (litigation records and NPE-directory designations) rather than on the assignment chain, which here is effectively a null result.

Conveyance types observed: none beyond the original corporate assignment. No Security Agreement, Merger, Change of Name, License, or Release recordings were surfaced for this number. No assignment correspondent recurs because there are no recorded assignments to compare.

Timeline diagram

timeline
    title Ownership of US 12554825
    2001 : Proxense founded in Bend Oregon
    2007 : Provisional application filed by Giobbi
    2008 : Non provisional application filed
    2015 : First continuation filed
    2016 : Entire portfolio offered for sale
    2019 : Second continuation filed
    2021 : Third continuation filed as 17394837
    2021 : First enforcement suit filed
    2025 : Google dispute settled
    2026 : Patent issued to Proxense LLC
    2026 : Intel dispute settled

NPE / troll-pattern signals

1. Shell-entity transfer — not present. There is no recorded transfer of this patent to any licensing LLC. The would-be shell step (the 2016 all-portfolio sale) never closed, and all litigation is brought in Proxense's own name. Note the inverse oddity: Proxense is itself a single-purpose, no-products entity, but it is the original assignee rather than a transferee shell — so this specific signal does not fire.
2. Known asserter in the chain — PRESENT (strong). The assignee is expressly labelled an NPE by two independent directories: Unified Patents — "The patents are owned and asserted by Proxense LLC, an NPE" (https://www.unifiedpatents.com/insights/2025/7/21/4000-for-proxense-authentication-patents-prior-art), and Stanford NPE Litigation Database — asserter type "Non Practicing Entity / 3 Failed startup" (https://npe.law.stanford.edu/case/286643). ip fray describes it as "a non-practising entity (NPE) and former operating company." Proxense is not on the classic Acacia/Marathon/IV list; it is a self-asserting, original-owner NPE.
3. Repeat correspondent across the chain — unclear / not applicable. With zero recorded assignments, there is no assignment correspondent series to test. The prosecution attorney of record for the family is Patent Law Works LLP (address of record on the sibling '954 application: "Patent Law Works/Proxense, 165 South Main St, Suite 2, Salt Lake City, UT 84111"). That is a prosecution firm, not a recording correspondent, and a single firm doing prosecution is not the signal you defined — flagging only for completeness.
4. Cascading transfers — not present. Ownership has been static at Proxense since inception; no chained-LLC sequence exists.
5. Pre-litigation transfer — not present for this patent. US 12,554,825 issued 2026‑02‑17, after every Proxense enforcement suit on record (Samsung 2021; Microsoft 2023; Apple/Intel 2024). There was no transfer, and the patent is too new to be the object of a venue/standing transfer. Its family siblings were asserted with no transfer at all.
6. Bankruptcy fire-sale — not present. No Chapter 7/11 filing surfaced for Proxense. The 2016 event was an out-of-court, unsolicited portfolio-sale solicitation, not a court-administered sale — worth logging as a distress signal, but not a bankruptcy fire-sale.
7. Privateering — not present. No operating company is using Proxense as a proxy; Proxense asserts its own original patents and markets them for sale. The direction of the 2016 letter (Proxense → operating companies) is the mirror image of privateering.
8. Defensive aggregator — not present (inverse confirmed). The chain does not terminate at RPX / AST / LOT / Unified / OIN. To the contrary, Unified Patents is running PATROLL prior-art bounties and IPRs against Proxense's patents — i.e., the defensive aggregators are arrayed against this patent family, not holding it. (Settlements exist with Samsung, Google (2025‑01‑24) and Intel (2026‑03‑24), but those are commercial licences, not defensive aggregation.)

Verdict

NPE — high confidence.

Justification: the patent never left its original owner, but that owner is a non-practicing entity in its own right — Proxense, LLC, is designated an NPE by Unified Patents ("owned and asserted by Proxense LLC, an NPE") and by Stanford's NPE database ("Non Practicing Entity — 3 Failed startup"), and on 2016‑07‑25 its counsel marketed the entire portfolio on the express representation that a buyer could "use the Proxense patents offensively against others" and that no licences existed. Proxense has since run a serial assertion campaign against Samsung, Google, LG, Hyundai, Microsoft, Apple and Intel with at least three settlements (Samsung, Google 2025‑01‑24, Intel 2026‑03‑24), while this specific number issued 2026‑02‑17 with an anticipated expiration of 2028‑12‑19 — a short enforcement window on a continuation that is itself an assertion-ready vehicle. Caveat on strength: because ownership never moved, the classic transfer-based signals (shell-entity transfer, cascading transfers, pre-litigation transfer, privateering) are all absent, so this verdict rests on the known-asserter signal plus documented failed-product/portfolio-liquidation evidence rather than on a reel/frame assignment chain — and no reel/frame citations exist to give, as no assignments are recorded.

Verify at: USPTO Assignment Center — https://assignmentcenter.uspto.gov/ (search "12554825") · Google Patents legal events — https://patents.google.com/patent/US12554825/en

I'll search for this specific patent number and its citations across USPTO and patent databases.

Initial searches didn't surface the specific number. Let me try more targeted sources.

Bottom line

I could not retrieve a "References Cited" (patent citations) list for US 12,554,825 B2 from any source available to me, and — critically — the authoritative full text supplied for this patent does not contain that section. It was truncated while describing FIG. 11D (step 1150) and therefore ends before both the claims and the "References Cited" table. My searches for the literal number (see below) also returned no page carrying the '825 citation list.

I will not fabricate a citation list or a § 102 anticipation mapping. Doing so would require inventing both the examiner-cited references and the claim language they are mapped to — neither of which I have.

1. Search results for the literal number US 12,554,825

I searched, literally, for 12554825, "12,554,825", "US12554825 B2", and US12554825B2 (Google Patents), plus the application number 17/394,837.

• Zero direct hits. No index I reached returned a page for this patent number, its citation table, or its claims. This is consistent with the previous section's finding and I did not substitute a sibling number (per your strict rule).
• Results that did return were unrelated patents sharing digits (e.g., EP 3,929,229; US 7,452,525; AU 2022382981) or Proxense-family material on other patent numbers. None is US 12,554,825.
• Family/litigation results returned were for other Proxense patents — notably US 8,646,042 (IPR2024‑00573, Microsoft/Apple) and US 8,886,954 (the '954 Federal Circuit appeal). Those are different patents and cannot be attributed to the '825.

2. What the authoritative text does cite (the only "references" actually present)

The only references appearing anywhere in the supplied '825 text are four U.S. patent applications incorporated by reference by the specification (they appear twice — once in the FIG. 5 PDK discussion, once in the FIG. 6 reader discussion). Reproduced exactly as written, with the apparent internal inconsistency in the first entry flagged rather than corrected:

Important caveats you should apply to all four:

• These are incorporation-by-reference citations (foundation material for the PDK/RDC), not examiner citations in a "References Cited" list, and not asserted § 102 art.
• Three of the four (the '11/6xx,xxx and '11/744,xxx series) have 2007 filing dates on or near the '825's own 2007‑12‑19 priority date, so on their face they are at best § 102(a)/(e)-type candidates and would face the same priority-date problems a defendant would litigate; I have not verified their publication or issue dates, so I make no anticipation assertion for any of them.

The Google Patents page also lists "Prior art keywords": computing device, pdk, reader, login information, access, and Prior art date 2007‑12‑19. Those keyword tags are algorithmic, not a citation list.

3. What is missing, and why the § 102 question cannot be answered as posed

To answer "which claim(s) it potentially anticipates under § 102," I need two things, and I have neither:

1. The '825 claim set — the supplied text stops mid‑sentence in FIG. 11D and contains no claims at all. Anticipation is an element-by-element comparison to a specific claim; without the claims, any mapping is invented.
2. The '825 "References Cited" list — absent from the supplied text and not surfaced by any search.

I also note, as did the earlier section, an unresolved date discrepancy: your instruction says 2026‑04‑26, while the authoritative source fetch timestamp is 2026‑09‑30. This does not affect the citation question but should be reconciled before relying on any date.

4. What I can defensibly offer instead

• A proxy for the claim set, clearly labeled as a proxy: the two granted siblings in the same chain — US 10,469,456 B1 (app 14/973,565, 18 claims) and US 11,086,979 B1 (app 16/672,214, 20 claims) — share this specification. Their claims would let me show the shape of the likely '825 claims, but they are not the '825 claims and must not be treated as such.
• The real next step for an authoritative answer: pull the printed patent or the USPTO Patent Public Search (patft) / PatentCenter record for application 17/394,837. That record holds the actual "References Cited" (both U.S. and foreign), the examiner's § 102/§ 103 rejections, and the issued claims — the three inputs needed for the exercise you asked for. I could not reach those pages from the search tools available here.
• Context on the family's known prior art (flagging hard that these belong to other Proxense patents and must not be attributed to the '825): the IPR2024‑00573 record concerns US 8,646,042; the '954 appeal concerns US 8,886,954. I saw no reference — e.g., no Gale, no Bishop, no specific art — tied to the '825.

5. Explicit uncertainty statement

• High confidence: US 12,554,825 B2 exists in the family; app 17/394,837; the only references named in its supplied text are the four incorporated applications above; the supplied text contains no claims and no "References Cited" section.
• None / unknown: the examiner-cited prior art for the '825; the '825 claim count and language; and therefore any § 102 anticipation mapping. I am stating this as unknown rather than guessing, per your instruction to say so explicitly.

If you can paste the References Cited section and the claims (or authorize a proxy analysis using US 11,086,979 B1 / US 10,469,456 B1), I can complete the per‑reference citation table and the § 102 mapping properly.

I'll build on the prior sections. First, let me try to retrieve the "Prior Art" citations and claim text for this patent, since the supplied page text was truncated before both the claims and any references-cited table.

Let me try to retrieve the actual claims and the "References Cited" list for the '825 patent and its immediate parents.

Bottom line up front

I can give you a disclosure-level § 103 attack map, and a genuinely useful one — but I must flag two hard limits before you rely on it:

1. The "Prior Art" section of the supplied page contains no references. Google Patents renders of this type normally include "References Cited" (patent citations, non‑patent citations) and a "Cited By" table. In the authoritative text I was handed, the entire Prior Art content is: Prior art keywords = {computing device, pdk, reader, login information, access} and Prior art date = 2007‑12‑19. There is no citation table, and the text is truncated mid‑sentence in the FIG. 11D description of step 1150 — i.e., before the claims. So an instruction to "use the results from the Prior Art section" cannot be executed literally; I reconstructed what I could from searches and label every reconstructed item below.

2. Claim-level § 103 analysis is not possible yet. The previous section already flagged that the issued claims of US 12,554,825 are unavailable in the material I have. That remains true. What follows is an obviousness analysis of the disclosed subject matter — which is the correct starting point for building a claim chart, but is not a claim-by-claim opinion.

Two cross-reference notes: (a) the April 26, 2026 / 2026‑09‑30 date discrepancy flagged in the prior section is still unresolved; (b) my searches surfaced additional Proxense PTAB matters not in the prior section's docket list — IPR2024‑00405 (Microsoft v. Proxense, US 10,073,960), IPR2024‑01320 (Google v. Proxense, US 10,073,960), and IPR2025‑00562 (Apple v. Proxense, US 10,071,188). That refines rather than contradicts the earlier docket summary.

1. The gating question that decides the whole § 103 analysis: the effective filing date

This is not a formality here — it determines which prior art exists at all, and it determines whether Proxense's own family is available against it.

• The '825 (app. 17/394,837, filed 2021‑08‑05) claims a chain: 16/672,214 (2019) ← 14/973,565 (2015) ← 12/340,501 (2008‑12‑19) ← provisional 61/015,110 (2007‑12‑19).
• Under AIA § 3(n)(1), pre‑AIA §§ 102/103 govern only if the application never contained a claim with an effective filing date on or after 2013‑03‑16. If any claim (now or ever) depends on subject matter not supported by the 2007 provisional or 2008 non‑provisional, the AIA regime applies and each claim's date must be fixed individually (§ 112(a) support; Dynamic Drinkware v. National Graphics, 800 F.3d 1375 (Fed. Cir. 2015)).

Why it matters, concretely:

Key strategic insight: Giobbi‑'139 and Giobbi‑'157 — the exact combination petitioners successfully used in the Proxense IPRs — become available only if the '825's claims slip past the pre‑2013 window. For a 2007/2008 date, they are § 102(e) art at best and are knocked out for § 103 by pre‑AIA § 103(c) common ownership (both Proxense). That is the single highest-leverage issue in this analysis. The family's continued new‑matter filings (including the 2025 child 19/030,369 → US 2025/0165580 A1) make that slip non-trivial to rule out.

Also note the specification's internal inconsistency: it incorporates by reference "U.S. patent application Ser. No. 12/292,330, filed Nov. 30, 2005." Per your instruction I interpret that literally — but the 12/292,xxx series did not exist in 2005, and the sibling '483 patent (Giobbi, Personal Digital Key and Receiver/Decoder Circuit System and Method) shows a § 65 prior publication of US 2006/0136742 A1 (Jun. 22, 2006), consistent with a Dec. 2005 filing. A defective incorporation by reference is worth a § 112/priority look; it cannot create an earlier priority date in any event.

2. Prior art I could actually recover (labeled by confidence)

Recovered from a Google Patents citation table that lists US12554825B2 among its documents (asterisked entries, i.e., citations-of-record convention). ⚠️ I could not confirm which patent's citation table this is, nor that these were cited against the '825:

Recovered from the family prosecution record (Justia listing associated with US 11,086,979 B1, the immediate parent): a long § 102 citation list of 2001–2003 US publications — e.g., US 2002/0174348 (Ting); 2002/0177460 (Beasley); 2002/0178063 (Gravelle); 2002/0187746 (Cheng); 2002/0191816 (Maritzen); 2003/0034877 (Miller); 2003/0061172 (Robinson); 2003/0079133 (Breiter); 2003/0088441 (McNerney); 2003/0105719 (Berger); 2003/0109274 (Budka); and US 2003/0115351 (Giobbi). I have not verified the disclosures of these references and will not characterize them beyond their bibliographic entries.

References used against Proxense in the PTAB (Giobbi‑'157, Giobbi‑'139, Dua US 9,042,819, Broadcom EP 1536306 A1) — subject to the availability caveat in § 1.

From the patent's own text (usable as admissions): the Background admits users maintain "as many as a hundred of third‑party systems," that existing file‑encryption systems fail because of "administrative overhead," and the Detailed Description states: "In contrast to the prior art that requires that he be placed within 10 or less inches of the reader…" These are applicant admissions that proximity‑token access control existed in the field, with the only asserted difference being range.

3. Disclosure-to-art mapping (the chart foundation)

4. Proposed § 103 grounds, with KSR motivations

Framework: pre‑AIA § 103(a) as construed by KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398 (2007) and Graham v. John Deere, 383 U.S. 1 (1966). KSR was decided 2007‑04‑30 — before the asserted priority date — and holds that "the combination of familiar elements according to known methods is likely to be obvious when it does no more than yield predictable results."

Ground 1 — Microsoft '892 in view of a credential‑vault/auto‑fill reference; optionally further in view of Thomson EP 2104053 or Safend '868.
Microsoft's Proximity authentication supplies presence‑based authentication of a computing device to a wireless device. A password‑manager/auto‑fill reference supplies vaulted credential storage and automatic population/submission of sign‑on fields. Safend/Thomson supply protecting stored data and keys with encryption.
Motivation: the '825's own Background frames the problem — password sprawl across many third‑party systems and the burden of prior encryption tools. Combining proximity detection with a credential store yields the predictable benefit of no manual entry while retaining presence‑based security. Each element performs its own established function (In re Keller, 642 F.2d 413, 425 (CCPA 1981)).
Caveat worth flagging to any client: Microsoft's filing date (2008‑01‑09) postdates the 2007‑12‑19 priority date. Ground 1 lives or dies on the priority determination in § 1. It also needs a Dynamic Drinkware/Giacomini check on whether '892 claims benefit of an earlier provisional or CIP.

Ground 2 — pre‑2007 proximity/badge computer‑access system + pre‑2007 password vault/auto‑fill + pre‑2007 encrypted key store.
This is the date‑proof ground: it works regardless of which effective date wins. Motivation is the same, plus the applicant's own admissions that the prior art required ≤10 inches of token proximity. The asserted difference is range (10 inches → ~5 meters). Range is a scalable, routinely‑varied parameter, and the specification itself says the range "is completely scalable." Absent unexpected results, changing a known parameter is a design choice. Note the tradeoff: a longer range weakens presence‑as‑authentication, which the spec itself addresses via timed reconfirmation and biometrics — both known.

Ground 3 — Giobbi '139 in view of Giobbi '157 (the PTAB combination), further in view of a vault/auto‑fill reference.
Giobbi‑'139 teaches integrating reader functionality and enabling/disabling device functions based on PDK communication; Giobbi‑'157 teaches the PDK's tamper‑proof secure memory. Motivations: same inventor, same architecture, compatible components, predictable unification.
Availability caveat, and it is decisive: for a 2007/2008 effective date these are § 102(e) art subject to pre‑AIA § 103(c) (common ownership by Proxense) — unusable. They become § 102(b) art only if the effective date is on/after ~Oct. 2008 (as to '157) / Dec. 2005 (as to '139), which in practice means the AIA/post‑2013 timeline — precisely how petitioners used them against the '960/'188 patents.

Ground 4 — hierarchical item setup records over ACL inheritance / Group Policy registry policy stores. Motivation: avoid re‑specifying policy per file; inheritance is the known mechanism.

Ground 5 — duration timer and auto‑secure on exit over screensaver‑password/idle‑logout timers and proximity‑loss lockouts. Motivation: avoid false locks on momentary absence; a grace period is a routine design choice.

Ground 6 — master/proxy PDKs and backup password over RBAC and delegate‑credential systems.

5. Rebuttals and secondary considerations Proxense will raise

• Availability/priority (strongest defense): win the 2007/2008 date and Grounds 1, 3, and the vault‑storage references largely evaporate; § 103(c)/§ 102(b)(2)(C) removes the family art from Grounds 1/3.
• Different fields of endeavor: physical access control vs. web single‑sign‑on vs. file encryption — no motivation to combine. KSR and MPEP § 2143 cut against rigid field limits where the references address the same underlying problem (secure access to computing resources).
• Teaching away / change in principle of operation (In re Ratti, 270 F.2d 810 (CCPA 1959)): not obviously supported on this record.
• Nonobviousness evidence: long‑felt need, failure of others, industry praise, commercial success — but nexus must be shown (WBIP v. Kohler, 829 F.3d 1317 (Fed. Cir. 2016); In re GPAC, 57 F.3d 1573 (Fed. Cir. 1995)). Watch the double edge: a 19‑year alleged long‑felt need is undercut if Microsoft filed a proximity‑authentication application on 2008‑01‑09 — three weeks after the Proxense provisional. Near‑simultaneous independent invention by a third party is classic evidence that the solution was obvious to skilled artisans at the time.
• Hindsight: the examiner must consider each reference for what it actually teaches, not for the pieces that fit.

6. What would defeat the attack — and the two things the file history hints at

Two signals in the text you gave me are worth acting on:

1. The rescission-of-disclaimer paragraph. The '825 contains the boilerplate "Applicants hereby notify the USPTO that the claims of the present application are different from those of the parent application… Applicants rescind any disclaimer of claim scope made in the parent application." This indicates earlier prosecution statements narrowing scope — i.e., the applicant amended claims to overcome art somewhere in the chain. Locate that amendment. It will tell you exactly which reference the applicant was worried about, and whether the rescission is effective (it does not cure prosecution history estoppel already created in an issued parent).
2. The continuation pattern. Same disclosure, three issued patents ('456, '979, '825) plus a 2025 child. That pattern implies the '825's claims were drafted to be patentably distinct from the parents' claim sets, so the incremental limitation is where the fight will be — likely in the vault/PDK division of storage ("maximum mobility"), the detection‑engine event taxonomy, or the assigned‑ID grouping. Grounds 4–7 above target those increments.

7. What I need to make this a real opinion

1. The issued claims of US 12,554,825 (USPTO PatentCenter / Patent Public Search) — the single blocking gap.
2. The '825's own "References Cited" table and IDS/1449 — the page's Prior Art section lacked it.
3. The 12/340,501 specification and provisional 61/015,110 — to run the § 112(a) priority test per claim (this decides whether Giobbi '139/'157, Microsoft '892, EP 2104053, and Safend '868 are even available).
4. The prosecution histories of 14/973,565 and 16/672,214 — to find the amendment the rescission paragraph alludes to.

Confidence: High on the bibliographic facts, the priority‑date framework, and the § 103 legal scaffolding. Medium on the relevance of the recovered Microsoft/Thomson/Safend/F5 citations (I could not confirm they are the '825's citations of record). Low/unknown on the actual claims — every ground above is keyed to the disclosure, not to claim language, and should be re‑charted once the claims are in hand.