US 10717446

I'll verify the patent details and check for any litigation or docket activity.

Let me get the specific claim language and check whether this exact patent is at issue in the identified litigation.

Let me retrieve the actual claim set of the '446 patent.

US 10,717,446 B2 — Verification Summary

I searched for the literal identifier 10717446 and excluded similar numbers. Findings below; where sources conflict I say so rather than resolving it silently.

Bibliographic data (confirmed)

Source: Google Patents US10717446B2; Unified Patents portal

Ownership chain (per the assignment record): IBM → Daedalus Group LLC (recorded 2019-11-14 / 2020-01-27) → Slingshot IoT LLC (2020-01-29) → Granite Vehicle Ventures LLC (2023-07-27). Note a metadata quirk: the same Google Patents page labels both "Original Assignee: Slingshot Iot LLC" and an earlier "Assigned to International Business Machines Corporation" entry; the assignment documents show IBM as the original assignee.

Abstract (verbatim)

"A computer-implemented method, system, and/or computer program product controls a driving mode of a self-driving vehicle (SDV). One or more processors compare a control processor competence level of an on-board SDV control processor in controlling the SDV to a human driver competence level of a human driver in controlling the SDV while the SDV encounters a current roadway condition which is a result of current weather conditions of the roadway on which the SDV is currently traveling. One or more processors then selectively assign control of the SDV to the SDV control processor or to the human driver while the SDV encounters the current roadway condition based on which of the control processor competence level and the human driver competence level is relatively higher to one another."

Independent claims — plain language

Caveat on precision: the authoritative full text furnished to me ends inside the FIG. 4 description and does not include the numbered claim set, and my searches did not return verbatim claim text for the '446 patent specifically (the claim-language hits I found belong to sibling patents '765, '402 and '004). Accordingly, the following is a plain-language reconstruction from the '446 SUMMARY section — which in this family mirrors claim 1 nearly word-for-word — not a verbatim reproduction. Treat exact wording as unverified.

The '446 is a continuation whose three statutory classes of independent claim (method, system, computer program product) all recite the same core logic:

1. Determine a "control processor competence level" of the on-board SDV control processor — how competent that processor is at driving the SDV while the SDV encounters the current roadway condition, where that condition results from current weather conditions of the roadway being traveled.
2. Receive a driver profile describing a "human driver competence level" — how competent the human is at manually controlling the SDV under that same weather-derived roadway condition.
3. Compare the two competence levels.
4. Selectively assign control of the SDV to the SDV control processor (autonomous mode) or to the human driver (manual mode) based on which competence level is relatively higher, while the SDV encounters that condition.

The weather-condition limitation is the apparent point of novelty relative to the earlier members of the chain (which used fault/anomaly triggers). Dependent claims in this family add variants such as history-based vs. capability-based competence determination, weighting/summing of control factors, cohort-based analysis, roadway-width inputs, and a minimum-competence threshold that causes the SDV to pull over if neither operator is competent.

Related family members (same specification, Google Patents family): US 9,566,986 (14/865,393, 2015-09-25); US 10,029,701 (15/341,225); US 10,717,446 (15/955,874); US 11,091,171 (16/997,202); US 11,597,402; US 11,738,765 (17/374,656); US 12,037,004 (18/222,774); plus pending US 2025/0018963 A1.

Litigation / 2026 docket check — important negative finding

I found no court docket or PTAB proceeding in which US 10,717,446 itself is the asserted or challenged patent. The activity I located involves sibling continuation patents with different numbers, and I am flagging this explicitly so the '446 is not conflated with them:

• Granite Vehicle Ventures LLC v. Tesla, Inc., No. 2:24-cv-01007-JRG (E.D. Tex., filed 2024-12-06). Granite's P.R. 3-1 disclosure asserts US 11,597,402, US 11,738,765, and US 12,037,004 — not 10,717,446. Accused products are Tesla FSD vehicles (Models S/3/X/Y, Cybertruck). (P.R. 3-1 contentions; Unified Patents summary)
• PTAB IPRs filed by Tesla against Granite: IPR2025-00943 ('402), IPR2025-01034 ('004), and a parallel '765 proceeding. In at least two of these, Patent Owner filed statutory disclaimers (Aug 8, 2025 and Aug 22, 2025) — again directed to the asserted patents, not the '446.
• In re Tesla, Inc., No. 2026-116 (Fed. Cir. 2026-02-27) — appeal of the PTAB's discretionary denial of institution, decided on non-merits/§314(d) grounds.
• Tesla, Inc. v. Granite Vehicle Ventures LLC, Supreme Court No. 26-136 (petition filed 2026-07-27; Unified Patents amicus brief filed 2026-08-25) — cert. petition from that Federal Circuit decision, again about IPR institution timing, not the '446's validity.

None of the above names US 10,717,446. If you need litigation specifically touching the '446, I do not have authoritative evidence of any, and I would not assert there is none.

Stated uncertainties and conflicts

1. Verbatim claims not verified — see caveat above.
2. Legal status conflict. The authoritative Google Patents text I was given lists the status as "Expired – Fee Related, expires 2035-10-25," while third-party aggregators (OnScope, Unified Patents) show it "In Force"/"Active." Google's own snippet also shows "Active, expires." I cannot reconcile these from available sources; a maintenance-fee-lapse determination (or a late payment/petition to revive) would explain it, but I have not confirmed that.
3. "Original assignee" metadata conflict on the Google page (IBM vs. Slingshot IoT), as noted.
4. My search coverage of the USPTO and CAFC/PTAB sources was via public web indexes (Google Patents, USPTO PTAB/PTACTS, federal court filings mirrored online), not a direct PatentCenter or PACER query — so an unindexed or very recent filing on the '446 would not have surfaced.

I'll research litigation involving US patent 10717446. Let me search multiple sources.

Let me search for litigation involving the current assignee and the patent family.

Let me search specifically for the '446 patent in litigation contexts and check the Granite v. Tesla case assertions.

Summary Answer

No litigation specifically asserting U.S. Patent No. 10,717,446 has been found. I searched patent‑litigation sources (Unified Patents portal, RPX Empower, Docket Alarm/Justia dockets, Law360, PACER‑derived dockets, PTAB/PTActs filings, and the Federal Circuit/Supreme Court dockets) and could not identify any complaint, counterclaim, or PTAB proceeding in which the ’446 patent itself is the patent‑in‑suit or the challenged patent.

Importantly, I want to flag a potential confusion trap: the same owner (Granite Vehicle Ventures LLC) is currently litigating a highly publicized case against Tesla — but that case asserts sibling patents in the same family, not the ’446 patent. I am reporting it below strictly as family/owner context and explicitly noting that the ’446 is not among the asserted patents.

---

Litigation found — but NOT involving the ’446 patent (family/owner context only)

Granite Vehicle Ventures LLC v. Tesla, Inc.

• Plaintiff: Granite Vehicle Ventures LLC (current assignee of the ’446 patent)
• Defendant: Tesla, Inc.
• Jurisdiction / Case numbers:
  • Original: U.S. District Court for the Eastern District of Texas, Marshall Division — Case No. 2:24-cv-01007-JRG (Judge Rodney Gilstrap), filed December 6, 2024.
  • Transferred: U.S. District Court for the Northern District of California — Case No. 3:26-cv-01457-TLT (TSH) (transfer order unsealed/confirmed late January 2026).
• Asserted patents: U.S. Patent Nos. 11,597,402, 11,738,765, and 12,037,004 — all continuations/descendants in the same specification family as the ’446 patent, but the ’446 patent is not asserted.
• Accused technology: Tesla’s FSD (Supervised) driver‑assistance program.
• Status (current as of the reported record): Tesla filed a motion to partially dismiss and a motion to transfer to California; the case was ultimately transferred to N.D. Cal. Tesla asserted invalidity counterclaims. Tesla also filed a motion to compel venue discovery, which was litigated in E.D. Tex. before transfer.

Related PTAB / appellate proceedings (same family, NOT the ’446 patent):

• IPR2025-00943, Tesla, Inc. v. Granite Vehicle Ventures LLC — challenges claims of U.S. 11,597,402 ("Fault Remediation for Self-Driving Vehicles"); petition filed May 5, 2025; discretionary denial (Sept. 26, 2025).
• Additional Tesla IPRs including IPR2025-01035 (Tesla v. GVV) against related family patents — also subject to discretionary denial.
• In re Tesla, Inc. (Fed. Cir.) — Tesla’s mandamus petition over the PTAB discretionary denials was denied (February 2026).
• Tesla, Inc. v. Granite Vehicle Ventures LLC — Petition for Writ of Certiorari to the U.S. Supreme Court (Docket No. 26‑136, filed May 2026), with industry amicus briefs filed August 2026 (e.g., Unified Patents, CCIA, Alliance for Automotive Innovation), addressing discretionary IPR denials under § 315(b).

---

Important caveats

1. The ’446 patent is not an asserted patent in any proceeding I could locate — the filings I reviewed assert the ’402, ’765, and ’004 patents. Under your instruction not to conflate similar numbers, the ’446 patent should be treated as not currently in suit.
2. The ’446 is in the same patent family as the asserted patents (all share the "Controlling Driving Modes of Self-Driving Vehicles" specification and the 2015‑09‑25 priority date, with inventors Michael S. Gordon, James R. Kozloski, Ashish Kundu, Peter K. Malkin, and Clifford A. Pickover). A future amended complaint could add it, but as of the current record it has not been asserted.
3. Verification limits: My conclusions rely on publicly indexed litigation records. A definitive negative requires a confirmatory search of PACER (all district courts, including the Delaware, E.D. Tex., W.D. Tex., and N.D. Cal. dockets for Granite Vehicle Ventures / Slingshot IOT / Daedalus Group / IBM) and the PTAB's Patent Trial & Appeal Case Tracking System. I could not confirm with 100% certainty that no obscure or recently filed complaint names the ’446 patent, but no such case appears in the indexed records.

Bottom line: Based on available sources, there is no known litigation specifically involving U.S. Patent No. 10,717,446. The prominent active patent litigation by its owner, Granite Vehicle Ventures, targets Tesla but asserts other patents in the family (’402, ’765, ’004), and the associated PTAB, Federal Circuit, and Supreme Court proceedings likewise concern those other patents.

I'll verify the canonical "no proceedings" signal against what's publicly searchable before finalizing.

Web search surfaced Tesla v. Granite PTAB activity and a Supreme Court petition — I need to pin down which patent(s) those cover before I can attribute anything to the '446.

Proceedings overview

Zero AIA trial proceedings on file against US 10,717,446. The USPTO Open Data Portal's structured PTAB proceedings block for this patent returns no IPR, PGR, or CBM — so there is no "claims invalidated" proceeding, no "patentee survived" proceeding, and no settlement or institution-denial to report on this patent. All claims of the '446 are untested at the PTAB. I ran independent web searches for any older or recently-filed proceeding the ODP may not have indexed ("10717446" + IPR/PTAB, petitioner-side and owner-side queries) and found none referencing the '446 as the challenged patent. The only hits tying "10717446" to PTAB-hosted documents were unrelated (a Brazilian court gazette entry and a Form 4 accession number) or family-adjacent filings inside Tesla's petitions (see below).

What does exist is a dense, well-documented PTAB/litigation war over the '446's patent-family siblings — US 11,597,402, US 11,738,765, and US 12,037,004 (all continuations claiming priority to the '446, all owned by Granite Vehicle Ventures LLC). Those four Tesla IPRs were all denied institution on discretionary grounds in 2025 and are now the subject of a live Supreme Court cert petition. That context matters to a defendant even though it is not, strictly, "PTAB proceedings on file" for this patent.

Bottom-line defensive posture: Neither template narrative fits. The patent is not "hardened by surviving IPRs" (it has never been instituted on), and no claim is dead (nothing has been canceled at the PTAB). The accurate statement is: the '446 is an untested, un-instituted patent whose commercial value rests on children that Tesla has already tried and failed to reach at the PTAB. The absence of IPR activity is not evidence of strength — it reflects the fact that the public challenge energy in this family has been aimed at the asserted continuations, plus discretionary-denial headwinds that have made institution extremely hard to obtain in 2025–2026.

---

Family-adjacent proceedings (NOT on the '446 — flagging so you aren't blindsided)

The following are real, verified proceedings, but they target the '446's children, not the '446. I am listing them because a defendant researching this patent family will encounter them, and because they establish the petitioner/owner/adjudicator pattern.

IPR2025-00943 — Tesla, Inc. v. Granite Vehicle Ventures LLC

• Type: Inter Partes Review
• Filed: 2025-05-05 (per third-party PTAB docket aggregation — treat as approximate)
• Patent challenged: US 11,597,402 (the "'402 Patent")
• Status: Discretionary Denial (institution denied 2025-09-26 by the Deputy Director under 35 U.S.C. § 314(a))
• Institution decision: Denied. The Director's delegate concluded "it is unlikely that a final written decision in [these] proceeding[s] will issue before the district court trial occurs," and that review would result in "significant duplication of effort, additional expense for the parties, and a risk of inconsistent decisions." (Quoted in In re Tesla, Inc., No. 2026-116, Fed. Cir. 2026-02-27, and incorporated into the later Director decision.)
• FWD: None — no trial instituted.
• Defensive value (family context): Patent Owner filed a statutory disclaimer of a large share of the '402 claims during the discretionary-denial fight. That tactic — disclaiming claims rather than defending them — is a signal about how confident the owner is in the challenged claim set.

IPR2025-00944 — Tesla, Inc. v. Granite Vehicle Ventures LLC

• Type: Inter Partes Review
• Filed: 2025-05 (approximate; same wave as -00943)
• Patent challenged: US 11,738,765 (the "'765 Patent") — mapping confirmed by the owner's document-labeling IPR2025-00944, EX2013 ('765 Patent disclaimer) in the -01035 record
• Status: Discretionary Denial (institution denied 2025-09-26, same Director decision as -00943)
• Institution decision: Denied under § 314(a) on the same time-to-trial rationale.
• FWD: None.
• Settlement/termination: Not a settlement — the petitions were denied; Tesla later sought fee refunds on the -01034/-01035 siblings.

IPR2025-01034 — Tesla, Inc. v. Granite Vehicle Ventures LLC

• Type: Inter Partes Review
• Filed: 2025-05 (wave filed after the -00943/-00944 group)
• Patent challenged: US 12,037,004 (the "'004 Patent")
• Status: Discretionary Denial
• Judge panel: Decision issued by Coke Morgan Stewart, Deputy Under Secretary of Commerce for Intellectual Property and Deputy Director of the USPTO (Director-level review, not a three-APJ merits panel)
• Institution decision: Denied 2025-10-10 under § 314(a). The Director expressly incorporated the analysis from IPR2025-00943 and -00944: "These cases present essentially the same discretionary considerations… Accordingly, that analysis is incorporated here."
• FWD: None.

IPR2025-01035 — Tesla, Inc. v. Granite Vehicle Ventures LLC

• Type: Inter Partes Review
• Filed: 2025-05-28 (per PTAB docket aggregation)
• Patent challenged: US 12,037,004 (second petition on the same patent as -01034)
• Status: Discretionary Denial; institution decision 2025-10-10; termination 2025-10-10
• Institution decision: Denied under § 314(a), consolidated into the same Director decision as -01034.
• Notable record: Tesla's papers argued Fintiv factor 3 (investment/claim disclaimers), attached exhibits on Gilstrap trial scheduling and time-to-trial statistics, and asserted the prior art case was "strong." The Director rejected the time-to-trial arguments holistically.
• Post-denial: Tesla filed requests for refund of the IPR request fees (2025-11-24 / 2025-12-04), approved 2025-11-26 — confirming no trial occurred.

Appeal: In re Tesla — family-level, not '446-specific

• Tesla petitioned the Federal Circuit for writ of mandamus challenging the non-institution decisions in IPR2025-00943, -00944, -01034, and -01035. Case No. 2026-116.
• Disposition: Denied (2026-02-27), nonprecedential order by Judges Taranto, Mayer, and Stark. The court held that institution decisions are committed to the Director's discretion and "final and nonappealable" under § 314(d), that mandamus is "ordinarily unavailable," and that Tesla's ultra vires theory presented "no colorable constitutional claim." Order: https://cafc.uscourts.gov/opinions-orders/26-116.ORDER.2-27-2026_2653825.pdf ; mirrored at https://www.courtlistener.com/opinion/[10801275](/patent/10801275)/in-re-tesla-inc/
• Supreme Court: Tesla filed a petition for certiorari (docketed as Tesla, Inc. v. Granite Vehicle Ventures LLC, likely No. 26-136; a 60-day extension pushed the filing deadline to 2026-07-27). Question presented: whether courts may review IPR denials resting on extra-statutory, time-to-trial grounds. Amicus briefs supporting Tesla were filed on 2026-08-28 by Unified Patents, LLC and by industry groups (CCIA, HTIA, Alliance for Automotive Innovation, National Retail Federation). https://www.supremecourt.gov/DocketPDF/26/26-136/[422322](/patent/422322)/20260828074131666_Certificate%20of%20Word%20Count.pdf

Underlying litigation (for context)

• Granite Vehicle Ventures LLC v. Tesla, Inc., originally E.D. Tex. No. 2:24-cv-01007 (Judge Gilstrap; complaint filed 2024-12-06; Tesla served 2024-12-10); the case was ordered transferred to N.D. Cal. in early 2026 (re-docketed around No. 3:26-cv-01457). The asserted patents in the PTAB record are the '402, '765, and '004 — not the '446. I could not confirm from the sources retrieved that the '446 itself has been asserted in that case; verify against the operative complaint before relying on it.

---

Strategic summary

Claim status on the '446. CANCELED: none — no claim of US 10,717,446 has ever been canceled, disclaimed, or held unpatentable in any AIA trial. SUSTAINED: none — no panel has ever upheld a claim of the '446. UNTESTED: all claims. The only claim-level attrition in this family is at the sibling level: Granite Vehicle Ventures filed statutory disclaimers of substantial portions of the '402 and '765 claim sets (and more of the '004 claims) during the 2025 discretionary-denial fights — i.e., the owner chose to narrow its own children rather than litigate their validity. That is a meaningful read-through on the family's confidence, but it is not a cancellation of any '446 claim, and you must not represent it as such.

Estoppel landscape — this is the good news for a defendant. Because no IPR was ever instituted on the '446, § 315(e)(2) estoppel has not attached against anyone with respect to this patent. Estoppel arises only after a final written decision; denied petitions generate none. Even as to the sibling patents, Tesla's discretionary denials leave Tesla free to press the same art as invalidity defenses in district court (subject to ordinary district-court limits, not § 315(e)). Practically: all prior-art grounds against the '446 remain available to any properly-positioned challenger. Two timing cautions: (1) § 315(b) — any defendant on the '446 must petition within one year of service of the first complaint asserting it, so identify the service date immediately; (2) if the '446 was asserted against Tesla in Dec-2024, Tesla's own § 315(b) window on the '446 has likely closed (around 2025-12-10), meaning a Tesla IPR on this specific patent is probably now time-barred — a fact worth confirming, since it would explain the absence of a Tesla petition on the '446.

Pattern signals. (a) One petitioner, four petitions, three patents, zero institutions — Tesla has shown sustained appetite to challenge this family, but every attempt died at the § 314(a) threshold, not on the merits. (b) A defensive aggregator is in the chain: Unified Patents ran a PATROLL crowdsourcing contest on US 12,037,004 (a '446 child, $2,000 prize, closed 2025-02-08) and filed an amicus brief supporting Tesla's cert petition. Unified has not (publicly) filed on the '446 — so there is no Unified-filed IPR to piggyback on, but there is free prior art and claim charts in the wild. (c) The patent owner is litigious and aggressive about forum (Texas filing, transfer fight, statutory disclaimers timed to defeat Fintiv), and Tesla has escalated hard (mandamus → SCOTUS cert). (d) The '446's status as a priority-root document means its specification is common to the asserted children — its disclosure will be picked apart in the children's cases, but its claims ride free.

---

Recommended next steps

1. Do not build a narrative on non-existent PTAB history. If opposing counsel or a demand letter implies the '446 "survived IPRs" or that its claims were "confirmed," that is unsupported — there is no proceeding, no FWD, no confirmation. Say so plainly and point them to the ODP record.
2. Pull the file wrapper and the family tree before filing anything. The '446 sits at the root of US 11,091,171, US 11,738,765, US 11,597,402, and US 12,037,004. Confirm in PatentCenter (a) the exact '446 claim set and (b) whether the '446 has been asserted against your client — the PTAB record for the siblings lists only '402/'765/'004.
3. Calendar § 315(b) now. Your one-year window runs from service of the first complaint asserting the '446. If you are outside that window, an IPR is off the table and your validity case is an E.D. Tex./N.D. Cal. invalidity case plus § 101/§ 112 challenges.
4. Reuse the Tesla/Unified record for the '446's siblings. Unified's PATROLL contest on US 12,037,004 (https://patroll.unifiedpatents.com/contests/G2ZqtvcKwR5WEPNab) and Tesla's IPR2025-01034/-01035 petitions and expert declarations against the '004 are the closest prior-art work product to the '446's disclosure. Because the '004 is a continuation of the '446, art that maps the '004's claims often maps the '446's — but check 2015-09-25 priority carefully (the '446's priority date is earlier than the children's), since art post-dating 2015-09-25 will not qualify against the '446.
5. Watch the cert petition. SCOTUS No. 26-136 (amicus briefing complete 2026-08-28) could reset the discretionary-denial standard. If the Court grants cert and Tesla prevails, the entire 2025–2026 wave of time-to-trial denials — including any future denial on the '446 — becomes reopenable or avoidable. Track the docket; if you have a live '446 petition strategy, consider whether a stay of any parallel litigation milestone is worth seeking pending that decision.
6. Verify the negative. The canonical ODP feed says zero proceedings on the '446 as of the most recent ingest. I found nothing contradicting it. If your timeline is high-stakes (e.g., an imminent § 315(b) deadline), re-run the PTAB E2E search (https://ptacts.uspto.gov/) and PatentCenter "Proceedings" tab the week you file, since petitions filed within the last few weeks may not yet be indexed.

Confidence and limits. I am confident about: zero PTAB proceedings on the '446 per the structured ODP data; the existence, patents, and discretionary-denial outcomes of Tesla's four family IPRs; the 2026-02-27 Federal Circuit mandamus denial in No. 2026-116; and the pending cert petition with Unified Patents' amicus support. I am less than fully confident about: the exact IPR filing dates (2025-05-05 / 2025-05-28 come from third-party aggregators, not the primary docket), the precise proceeding-number-to-patent mapping for -00943 and -00944 (inferred from the owner's exhibit labels in the -01035 record), the cert docket number (26-136 appears in a SCOTUS PDF filename but I did not verify the docket page), and whether the '446 itself is currently asserted against Tesla. Confirm those four items against primary sources before quoting them.

I'll research the assignment chain for US 10717446 using live sources, then ground the analysis in what's actually recorded.

Good initial data. Now let me pull the specific assignment records — reel/frame numbers, correspondents, and the entity principals.

Very useful — I've found reel/frame anchors. Let me pin down the full chain-of-title entries and the correspondents.

Assignment Record Reconstruction — US 10,717,446 B2

"Controlling driving modes of self-driving vehicles" · App. 15/955,874 · Filed 2018-04-18 · Issued 2020-07-21 · Priority 2015-09-25 (app. 14/865,393)

Source note / scope of verification. I reached the live web from this session but could not open the USPTO Assignment Center UI directly (it is not fetchable in this environment). Reel/frame numbers below are taken from USPTO PTAB filings for the sibling patents in this same family (IPR2025-00943 / -00944, Unified Patents / Tesla filings, and Petition 1557935), from IAM/Lexology and RPX/mondaq reporting on the IBM–Daedalus transaction, and from the Google Patents legal-events record for this patent. Where I could not verify a reel/frame or a correspondent for this specific patent, I say so explicitly rather than inferring it. All entries should be re-confirmed at https://assignmentcenter.uspto.gov/ by searching "10717446".

---

Inventors

All five are named on the 2015-09-25 priority application 14/865,393 and on the 2018-04-18 continuation 15/955,874, and are listed with Yorktown Heights, NY addresses in the USPTO application data for the later continuation 18/222,774 (PTAB Petition 1557935). The inventors executed an assignment to International Business Machines Corporation dated 2015-09-25, as recited in the Granite v. Tesla complaint (reported by IAM-adjacent press and by MLex, 2024-12-06).

Unusual-pattern check: inconclusive / no evidence found. I found no source establishing that any of the five departed IBM within 12 months of the 2015-09-25 filing. Note that this is a five-inventor, large-team IBM Research filing; the assignment to IBM is a routine employment-obligation assignment, not a distressed-inventor signature. I decline to assert departures I did not verify.

---

Original assignee

Two different facts are commonly conflated here, and they diverge:

• Assignee at filing (2015-09-25 / 2018-04-18): International Business Machines Corporation, Armonk, NY. IBM records the assignment from the inventors and appears in the Google Patents legal-events feed as assignee on 2018-04-18.
• Assignee named on the face of the issued patent (per Google Patents bibliographic "Original Assignee"): Slingshot IOT LLC. By the 2020-07-21 issue date the patent had already passed IBM → Daedalus Group → Slingshot IOT, and the printed "Slingshot Iot Llc" string is corroborated by the citation table for the sibling patents (e.g., US11738765B2's citation entry: "US10717446B2 … 2020-07-21 | Slingshot Iot Llc").

IBM — global IT services/hardware/software and one of the largest US patent holders (~60,000 US patents per the IAM/ktMINE US Patent 100 cited by IAM). It absolutely did ship products embodying self-driving/vehicle-control technology in the sense of research-stage work, but the claimed subject matter here (competence-level arbitration between an SDV control processor and a human driver) was never a commercial IBM product; it is IBM Research output. IBM's status: operating, solvent, publicly traded (NYSE: IBM). The patents were not sold in distress — IBM runs a standing patent-disposal program, and the September 2019 tranche was brokered with Houlihan Lokey (IAM/Lexology, 2019-11-22).

Slingshot IOT LLC — Maryland entity; per RPX/mondaq reporting it is one of a "web of NPEs" managed by York Eggleston and Keith Machen, related to Slingshot Technologies LLC and named on a "slingshot and stones / David and Goliath" theme. No products; licensing/assertion only.

---

Assignment timeline

Reel/frame reliability flag: Reels 051032/0784, 051710/0445 and 051733/0463 are documented in USPTO filings that concern other patents from the same IBM–Daedalus tranche (notably app. 16/172,085 / US 10,728,382, and the '402 IPR chain-of-title). Because the 2019 IBM→Daedalus transfer was a bulk conveyance of 500+ US assets in single transaction documents, the same reels cover this patent — but I have not independently read the reel index row for US 10,717,446 and am flagging that accordingly rather than presenting it as read.

1. 2015-09-25 (executed) / recorded on or about 2018-04-18 — Reel/frame not verified

• Conveyance: Assignment
• Assignor: Michael S. Gordon; James R. Kozloski; Ashish Kundu; Peter K. Malkin; Clifford A. Pickover
• Assignee: International Business Machines Corporation
• Correspondent: Not captured in retrieved sources. (IBM employment agreements of this vintage were typically recorded through IBM's own IP law department / outside counsel; do not treat this as a finding.)
• Context: Ordinary employment-obligation assignment of the priority application 14/865,393; the Google Patents feed carries an IBM assignment event dated 2018-04-18, matching the filing date of the continuation that issued as this patent.

2. 2019-09-30 (executed, effective 11:59 pm US Eastern) / recorded 2019-11-14 — Reel 051032/0784 (parallel record in the same tranche; see flag above)

• Conveyance: Assignment (Patent Assignment Agreement Ref. No. 1.190320)
• Assignor: International Business Machines Corporation
• Assignee: Daedalus Group, LLC (51 Pondfield Road, Suite 3, Bronxville, NY 10708 — address shared with Daedalus Blue LLC and Ginegar LLC)
• Correspondent: Not captured. Flag: this recording is itself the subject of a USPTO §1.182 petition to expunge, filed by Daedalus twelve days later (2019-11-26), on the basis that the full transaction agreement (containing §5.9 confidentiality terms) was recorded by mistake rather than the short-form assignment. That is a distinctive, checkable docket event.
• Context: Bulk sale of 500+ US assets and foreign counterparts for ~$9 million (largest tranche), brokered by Houlihan Lokey — a monetization-vehicle divestiture, not a distress sale. Daedalus managing principal: Ed Gomez (ex-Altitude Capital, ex-Walker Innovation, ex-General Patent Corporation).

3. 2020-01-27 (recorded) — Reel 051710/0445

• Conveyance: Assignment (corrected/replacement recording of the same IBM→Daedalus conveyance)
• Assignor: International Business Machines Corporation
• Assignee: Daedalus Group, LLC
• Correspondent: Not captured. This is the same record event that Google Patents renders as the 2020-01-27 "Daedalus Group, LLC" reassignment entry for this patent.
• Context: Administrative correction/re-recordation of #2 — not a new economic transfer.

4. 2020-01-29 (executed/recorded) — Reel 051733/0463

• Conveyance: Assignment
• Assignor: Daedalus Group, LLC
• Assignee: Slingshot IOT LLC (Maryland)
• Correspondent: Not captured. Flag for verification: this is the single highest-value correspondent to pull, because it is the first hop into the Eggleston/Machen entity web.
• Context: Downstream distribution — Daedalus handed ~75 assets to Slingshot IOT, which in turn pushed roughly a third onward to Quartz Auto Technologies, LLC (also Maryland), which sued Uber and Lyft in Feb 2020 (RPX/mondaq, 2020-07-22).

5. 2023-07-27 (executed/recorded) — Reel/frame not verified

• Conveyance: Assignment
• Assignor: Slingshot IOT LLC
• Assignee: Granite Vehicle Ventures LLC (Carrollton, TX per USPTO correspondence data; press describes it as Baltimore-based). Corporate parentage disclosed under Fed. R. Civ. P. 7.1: YEIP Holdings LLC and Albany Road Holdings LLC.
• Correspondent: Not verified as the assignment correspondent. Separately, Kinney & Lange, P.A., 333 S. 7th St., Suite 2700, Minneapolis appears as the correspondence address of record in the prosecution of Granite's continuation 18/222,774 (PTAB Petition 1557935), and Jan D. MacKinnon signed the assignee's power of attorney and the chain-of-title statement in the '402 IPR papers. I cannot confirm these are the same people/firm that recorded the 2023 assignment — flagging as a lead, not a finding.
• Context: Transfer to the asserting vehicle. Note the sequencing: the continuation 18/222,774 (which issued as US 12,037,004) was filed 2023-07-17 naming Granite as applicant, and the Slingshot→Granite assignment is dated 2023-07-27 — i.e., the assignment post-dates the filing of the last continuation by ten days, consistent with a confirmatory assignment papering a pre-arranged transfer.

Post-assignment litigation/administrative events (not assignments): 2024-12-06 — Granite Vehicle Ventures LLC v. Tesla, Inc., No. 2:24-cv-01007-JRG (E.D. Tex.), asserting US 11,597,402, 11,738,765, 12,037,004. 2025 — Tesla IPR petitions IPR2025-00943 / -00944; case later transferred to N.D. Cal.; Unified Patents filed a SCOTUS amicus brief in Tesla, Inc. v. Granite Vehicle Ventures LLC, No. 26-136.

---

Timeline diagram

timeline
    title Ownership of US 10717446
    2015 : Priority application filed
         : Inventors assign rights to IBM
    2018 : Continuation filed by IBM
    2019 : IBM sells 500 plus assets to Daedalus
         : Recorded at Reel 051032 Frame 0784
    2020 : IBM to Daedalus record re-filed
         : Daedalus assigns to Slingshot IOT
    2023 : Slingshot IOT assigns to Granite
    2024 : Granite sues Tesla in E D Texas
    2025 : Tesla petitions for IPR

---

NPE / troll-pattern signals

1. Shell-entity transfer — PRESENT.
Hard evidence, not naming inference: (a) IBM→Daedalus Group, LLC executed 2019-09-30 / recorded 2019-11-14 at Reel 051032/0784, in which the assignee's recorded address — 51 Pondfield Road, Suite 3, Bronxville, NY 10708 — is the same address USPTO records show for Daedalus Blue LLC and for Ginegar LLC, an entity formed in California on 2020-05-05 that received 30 assets from Daedalus on 2020-06-02; (b) Daedalus Group, LLC was formed in Delaware in May 2019, three months before the purchase, and IAM described it flatly as "not an operating company"; (c) the terminal holder Granite Vehicle Ventures LLC has no products, is characterized by Unified Patents as "an NPE," and its only disclosed activity in this record is a patent suit; (d) the intermediate Slingshot IOT LLC (Reel 051733/0463) is a Maryland assertion vehicle tied by RPX public-records analysis to Eggleston/Machen.

2. Known asserter in the chain — PRESENT.
Not on the classic Acacia/Marathon/IV/Wi-LAN canonical list, but confirmed by the two neutral directories you asked me to cross-check:

• Unified Patents — "U.S. Patent 12,037,004, owned and asserted by Granite Vehicle Ventures LLC, an NPE" (PATROLL contest, ended Feb 2025); and Unified is amicus at the Supreme Court in Tesla v. Granite (No. 26-136).
• RPX — Daedalus Group LLC and its affiliate Daedalus Blue LLC are named in RPX coverage of the post-IBM assertion wave; Slingshot IOT LLC is identified as part of the Eggleston/Machen NPE web, which RPX tied to "more than a dozen patent assertion entities." Related former-IBM tranches from the same Daedalus pool went to K.Mizra/Ginegar, Quartz Auto Technologies (Uber and Lyft suits), Taasera Licensing (Quest), and IP Edge entities (Noblewood, Pixeltide Pathway, Wikeshire).
• Litigation-finance overlay: related Eggleston/Machen "Slingshot" entity litigation in Delaware records references Burford Capital as a funding source (sanctions briefing, D. Del.). This is a third-party-funding signal; note the entity-name overlap is Slingshot Technologies, not necessarily Slingshot IOT — I am flagging the reported common management, not asserting the same entity.

3. Repeat correspondent across the chain — UNCLEAR (not established).
I could not retrieve the correspondents of record for reels 051032/0784, 051710/0445, or 051733/0463. What I do have are two adjacent names — Kinney & Lange, P.A. (Minneapolis) as correspondence address on Granite's continuation prosecution, and Jan D. MacKinnon as the signer of Granite's POA / chain-of-title statement — but neither is confirmed as the assignment-recording correspondent, and a single appearance would not meet your recurrence threshold anyway. This is the one field worth pulling by hand at Assignment Center; if MacKinnon (or a single Bronxville/Maryland-area filer) also appears on the Daedalus→Slingshot record, signal 3 flips to PRESENT and would corroborate signal 4.

4. Cascading transfers — PRESENT.
Four recorded hops inside the tranche, with three of them inside a ~10-week window: 2019-11-14 (Reel 051032/0784) → 2020-01-27 (Reel 051710/0445) → 2020-01-29 (Reel 051733/0463, Daedalus→Slingshot IOT). Within the same period the same assignor sprayed subsets of the identical asset pool to Ginegar/K.Mizra (2020-06-02), Quartz Auto, and Taasera — i.e., Daedalus functioned as a distribution hub, which is the classic chained-LLC pattern. The final hop (Slingshot IOT → Granite, 2023-07-27) is separated by ~3.3 years, so the full chain is not compressed end-to-end.

5. Pre-litigation transfer — NOT PRESENT (as to this patent's terminal hop).
Slingshot IOT → Granite was executed/recorded 2023-07-27; the first infringement suit was filed 2024-12-06 — roughly 16 months, well outside the 6-month window. However, the sequence is still an arrangement-to-assert tell: the last asserted continuation was filed 2023-07-17 naming Granite as applicant and the assignment is dated 2023-07-27, and the asserted patents issued after the transfer, so Granite acquired applications and litigated the issued patents. Note one anomaly worth a query: Google Patents' legal-status field for US 10,717,446B2 itself reads "Expired – Fee Related" with adjusted expiration 2035-10-25 — if accurate, the '746 maintenance fee lapsed and Granite asserted only the later continuations ('402, '765, '004), which would be consistent with a family-management strategy of letting an unused member go. Flagged as unverified; verify at USPTO Patent Center.

6. Bankruptcy fire-sale — NOT PRESENT.
No bankruptcy. IBM is a solvent, currently operating NYSE-listed company; the 2019 conveyance was a negotiated ~$9 million portfolio sale advised by Houlihan Lokey. This is a monetization divestiture, not a Chapter 7/11 sale.

7. Privateering — NOT PRESENT (weak circumstantial only).
Classic privateering requires the operating company to direct or benefit from the NPE's assertions against its own competitors. Nothing in the record shows IBM coordinating Granite's Tesla suit, and Tesla is not an IBM competitor in enterprise IT. There is a notable timing adjacency — LOT Network itself publicized that IBM sold hundreds of patents to Daedalus Group "about two weeks before joining the LOT Network" — but that is a screening/timing observation, not evidence of privateering, and I am not treating it as a signal.

8. Defensive aggregator (anti-NPE) — NOT PRESENT.
The chain terminates at Granite Vehicle Ventures LLC, which is actively asserting. No RPX, AST, LOT, Unified Patents or OIN acquisition appears anywhere in the chain. Inverse check: Unified Patents is on the other side (PATROLL prior-art bounties on '004 and '402, amicus at SCOTUS), which confirms the patent is not neutralized.

---

Verdict

NPE — high confidence.

The chain of record is IBM → Daedalus Group, LLC (executed 2019-09-30, recorded Reel 051032/0784; re-recorded 2020-01-27 at Reel 051710/0445) → Slingshot IOT LLC (Reel 051733/0463, ~2020-01-29) → Granite Vehicle Ventures LLC (2023-07-27), a four-hop monetization cascade on ~$9M of IBM assets terminating in an entity that Unified Patents formally identifies as "an NPE" and that sued Tesla on 2024-12-06 (E.D. Tex. 2:24-cv-01007). Two independent strong signals are present — shell-entity transfer (Bronxville, NY assignee address shared across Daedalus, Daedalus Blue and Ginegar; a Maryland assertion vehicle; a Texas/Baltimore litigation shell with no products) and known asserter in the chain (Daedalus, Slingshot IOT and Granite all surfaced by Unified Patents/RPX, plus the same IBM tranche feeding Ginegar/K.Mizra, Quartz Auto, Taasera and IP Edge campaigns) — with a third, cascading transfers, corroborated by the 2019-11-14 → 2020-01-29 recording cluster. The one materially open item is signal 3 (repeat correspondent): the recording correspondents for reels 051032/0784, 051710/0445 and 051733/0463 are not in the sources I could reach, and pulling them is the single highest-value follow-up.

Verify at: https://assignmentcenter.uspto.gov/ (search patent number 10717446; also cross-check the family members 11,597,402 / 11,738,765 / 12,037,004, whose IPR papers carry the cleanest chain-of-title recitals). Companion indexes: Google Patents legal events for US10717446; USPTO Patent Center for app. 15/955,874; PTAB IPR2025-00943 and IPR2025-00944; Unified Patents patent portal entry for US-10717446-B2.

---

Caveats I want on the record: (i) reel/frame values marked "not verified" are drawn from sibling-patent filings for the same bulk transaction, which is strong but not identical to reading this patent's own reel row; (ii) I found no correspondent names for the assignment recordings themselves and deliberately did not guess; (iii) the "Expired – Fee Related" status for the '746 comes from Google Patents' automated legal-status field, which is an assumption on that page, not a USPTO determination I confirmed.

I'll research US patent 10717446 and its cited prior art. Let me search multiple sources.

Let me get the specific patent citation table (prior art cited in the patent) and the prosecution history.

I have substantial data. Let me verify the key cited references and the actual "References Cited" list for this patent.

Prior-Art Analysis — US 10,717,446 B2

1. Identification of the patent (interpreted literally)

Critical threshold point for §102: because the earliest effective priority is 2015‑09‑25, only references with a date before that (or earlier‑filed/earlier-published applications) can be prior art under 35 U.S.C. §102(a)/102(b)/102(e). Anything dated later that appears in the record is a "Cited By" / IDS entry, not prior art, and cannot anticipate.

Continuity chain (all in the same family — NOT prior art to one another):
14/865,393 (US 9,566,986) → 15/341,225 (US 10,029,701) → 15/955,874 (US 10,717,446) → 16/997,202 (US 11,091,171) → 17/374,656 (US 11,738,765) / 17/719,541 (US 11,597,402) / 18/222,774 (US 12,037,004). Family ID 57964734.

The independent claims under scrutiny are claim 1 (method), claim 10 (computer program product) and claim 20 (system). Claim 1 requires, in substance:

• (a) determining a control‑processor competence level of the SDV control processor in controlling the SDV while it encounters a current roadway condition that results from current weather conditions;
• (b) receiving a driver profile describing a human‑driver competence level in manually controlling the SDV in that same condition;
• (c) comparing the two competence levels; and
• (d) selectively assigning control to whichever is relatively higher.
  (Claim 20 is the same architecture but recites "current lighting conditions" instead of weather.)

---

2. Sources and important limitations (stated up front)

• I retrieved the granted claims, abstract, description, family and the "References Cited" listing from Google Patents and uspto.report (URLs above).
• I could not retrieve the patent's own prosecution‑specific "Patent Citations" table with certainty distinguishing applicant‑cited vs. examiner‑cited vs. third‑party "Cited By" items. The uspto.report front‑page list is captioned "References Cited [Referenced By]" and merges the two. Some entries (e.g., 2016–2018 items) are therefore almost certainly cited‑by rather than prior art.
• No element‑by‑element §102 mapping below is a legal conclusion. The Examiner allowed all 20 claims, which means the Examiner did not find any single one of these references anticipatory. The "potentially anticipates" column is my technical relevance assessment, and for every reference I flag the missing element(s) that prevented anticipation.
• Where I am not certain of a reference's precise content, I say so explicitly rather than inventing a disclosure.

---

3. Most relevant cited patent prior art (pre‑2015‑09‑25) and potential §102 relevance

Foreign patent documents cited (all pre‑2015, mostly peripheral): CN 1135063 (1996); CN 2349068 (1999); CN 201004265 (2008); CN 202012052 (2011); CN 202038228 (2011); CN 102650882 (2012); CN 202772924 (2013); CN 104900018 (2015); EP 0582236 (1994); WO 2014/058263; WO 2014/066721; WO 2014/147361; WO 2014/148975; WO 2014/148976; WO 2015/024616; WO 2015/056105; WO 2015/156146. None of these non‑English‑language references, on their faces/titles, discloses the processor‑competence‑vs‑driver‑competence arbitration that defines claim 1; they are best characterized as background in vehicle‑control and telematics.

---

4. Non‑patent literature cited ("Other References")

These were cited by the applicant and are informative but not anticipatory of the competence‑arbitration claims:

• P. Mell et al., The NIST Definition of Cloud Computing, NIST, Sep. 2011 — supports claim 19 (cloud‑delivered program code) as background.
• Anonymous, "Car Built‑In Mechanism to Enforce Mandatory Self‑Driving Mode," IP.com No. 000234916, Feb. 14, 2014 — this is the most substantively relevant NPL item; it is the closest NPL theme to the "prohibit manual mode / force autonomous" subject matter of dependent claims 6 and 15.
• Anonymous, "Diagnostics Mechanism for Self‑Driving Cars to Validate Self‑Driving Capabilities," IP.com No. 000237171, Jun. 6, 2014 — background for self‑diagnosis of SDV capability (relevant to dependent claim 9).
• Popular‑press and academic items (CNBC 2014; Bloomberg 2014; Forbes 2015; MIT Tech. Review 2014; Boing Boing 2015; Horberry et al. 2006; Vanderbilt, Wired 2012; etc.) — background/context only.

---

5. Items that are NOT prior art and should be excluded

• All same‑family members (US 9,566,986; US 10,029,701; US 11,091,171; US 11,738,765; US 11,597,402; US 12,037,004 and their publications) — common priority, not §102 art.
• Same‑assignee/IBM sibling patents filed with overlapping priority (US 9,834,224; US 9,781,145; US 9,944,291; US 10,543,844; etc.) — these share inventors/organisation and overlap in time; treat carefully and never as §102 art to 10,717,446 without checking the exact §102(e) date.
• Everything dated after 2015‑09‑25 appearing in the record (e.g., US 2016/0304122; US 2017/0010613; US 2018/0093631; US 9,628,975; US 9,718,468; US 9,881,427) is cited‑by material, not prior art.

---

6. Litigation / post‑grant context worth flagging

The public record shows this family has been the subject of IPR activity, including papers captioned "IPR2025‑00944" with a Tesla exhibit ("Tesla EX1002"), and a "Declaration of Christopher Wilson" directed to U.S. Patent No. 12,037,004 (a later family member), addressing claims reciting a weighted voting system with active‑learning‑based inputs. See the PTAB documents surfaced under ptacts.uspto.gov (petition files 1557830 / 1557935 / 1557934). I could not retrieve the specific prior‑art grounds and exhibits of that IPR within this session, so I cannot name the art the petitioner(s) assert as the "most relevant." If you need the definitive "most relevant prior art" set, the IPR petition(s) for this family are the single best source — I'd recommend pulling IPR2025‑00944 directly from PTAB.

---

7. Bottom line

1. The literal identifier 10717446 = US 10,717,446 B2, Controlling driving modes of self-driving vehicles (Gordon et al.), Slingshot IOT / Granite Vehicle Ventures.
2. Under §102, only references before the 2015‑09‑25 priority qualify. On the record, no single cited reference discloses all of claim 1's elements — specifically the pairing of (a) a computed control‑processor competence level with (b) a human‑driver competence level, (c) their comparison, and (d) assignment of control to the relatively higher one while the SDV encounters a weather‑derived roadway condition. That is precisely why the Examiner allowed the claims.
3. The closest cited prior art in technical terms is US 8,874,305 (Dolgov, Google) for driver‑state‑based behavior/hand‑off, with US 7,894,951 (Norris) and US 8,078,349 (Prada Gomez) supplying autonomous/manual transition concepts, and US 8,140,358 (Ling, Progressive) supplying driver‑profile competence/rating. Each lacks at least the processor‑competence determination and/or the two‑level comparison of claim 1.
4. Claim 1/10/20 are the independent claims to map against; dependent claims 4, 5, 13, 14 (cohort/accident‑record competence), 6, 15 (weighted summed sensor readings → prohibit manual mode, cf. the IP.com "Enforce Mandatory Self‑Driving Mode" NPL), and 7, 16 (neither‑competent → stop) are the ones most likely to be attacked with combinations rather than a single anticipatory reference.
5. Honest uncertainty: I have verified the patent, its claims, its family, and the composition of its face‑of‑record citation lists, but I have not verified the full text of every single cited reference, and I could not retrieve the IPR grounds. Any statement above about a specific reference's disclosure should be confirmed against that reference's own text before being relied on.

Would you like me to (a) pull each of the ~16 highest‑relevance references individually and build a true element‑by‑element §102 claim chart for claim 1, or (b) retrieve the IPR2025‑00944 petition to identify the art the challenger is actually asserting?

I'll research the prior art cited against this patent and the relevant technological field before building the §103 analysis.

I have the full claim set now. Let me pin down the closest family reference and key weather/competence prior art.

Let me verify titles and content of the most probative cited references so I don't misstate them.

Obviousness Analysis — US 10,717,446 B2 under 35 U.S.C. § 103

Analyst note on sources. This analysis uses the "Prior Art" material appearing on the US 10,717,446 page: (i) the "References Cited" / "Patent Art" listings (Google Patents, uspto.report, and the Unified Patents portal, which lists 217 items of patent art), (ii) the cited non-patent literature, and (iii) references I retrieved directly. I also flag below one reference I found that is not in that listing but is highly material.

Contradiction flagged with the prior sections. The earlier "Patent summary" section stated that the numbered claim set was unavailable and reconstructed claim 1 in plain language. I have now retrieved the verbatim claims (claims 1–20, as reprinted on uspto.report for the '446 grant). Claim 1 matches the SUMMARY text almost word-for-word, so the earlier reconstruction was substantively correct; the difference is that the verbatim text is now available and I use it below. Also note that a defect in the earlier summary's family table: the '446's own "Related U.S. Patent Documents" lists only 15/341,225 → 10,029,701 and 14/865,393 → 9,566,986, not the longer chain listed earlier.

---

1. The claim to be analyzed

Claim 1 (verbatim, reproduced from the printed grant) requires:

• (a) determining a control processor competence level (CPCL) of an SDV control processor, "wherein the CPCL … describes a competence level of the SDV control processor in controlling the SDV while the SDV encounters a current roadway condition … wherein the current roadway condition is a result of current weather conditions of the roadway";
• (b) receiving a driver profile describing a human driver competence level (HDCL) "in manually controlling the SDV while the SDV encounters the current roadway condition";
• (c) comparing CPCL to HDCL; and
• (d) selectively assigning control to the processor or the human "based on which of the CPCL and HDCL is relatively higher."

Claim 20 is the system claim and substitutes "current lighting conditions" for weather. Dependent claims 2–9 add: on-board vs. remote assignment (2, 3); driver-cohort accident records (4); processor-cohort accident records (5); weighted-summed anomaly sensor readings that prohibit manual mode (6); a minimum competence threshold that, if unmet by both, pulls the SDV to a stop (7); roadway width as a further input (8); and routing to a resource provider (9).

Proposed level of ordinary skill (POSITA): a bachelor's degree in electrical, mechanical, or computer engineering (or equivalent) plus about 2–3 years' experience in ADAS/vehicle automation, or a master's degree with less industry experience. I offer this as a proposed construction, not as a finding of record.

Effective filing date is the pivotal fact. The '446 is a continuation claiming benefit of 14/865,393 (filed 2015-09-25, now US 9,566,986). If claim 1's weather limitation is supported by that parent, the effective filing date is 2015-09-25; if not, it is 2018-04-18. Almost every prior-art conclusion below changes depending on that answer, so I treat it as a fork rather than assume a value.

---

2. The single most material reference — and the biggest caveat

US 9,785,145 B2 / US 2017/0038773 A1, Controlling driving modes of self-driving vehicles, application 14/820,620, filed 2015-08-07, granted 2017-10-10, inventors Michael S. Gordon, James R. Kozloski, Ashish Kundu, Peter K. Malkin, Clifford A. Pickover — the same five inventors and the same assignee (IBM), but a different application and a different priority line from the '446. Its abstract and claim 1 disclose exactly:

"compare a control processor competence level of the on-board SDV control processor … to a human driver competence level of a human driver in controlling the SDV under the current condition of the roadway … then selectively assign control of the SDV to the on-board SDV control processor or to the human driver based on which of the control processor competence level and the human driver competence level is relatively higher to the other."

That is claim 1(a)–(d) of the '446 minus only the weather clause. Sources: US9785145 full text (PDF), Google Patents US9785145B2.

Two consequences, both important:

1. If the '446's effective filing date is 2018-04-18, then US 9,785,145 (granted 2017-10-10) and US 2017/0038773 A1 (published 2017-02-09) are §102(a)(1) prior art, and the common-ownership exception in §102(b)(2)(C) is unavailable against §102(a)(1) art. Claim 1 would then be squarely obvious (and plausibly anticipated) over this one reference, because the reference's own specification describes roadway conditions determined from "an information service (e.g., a weather station)" and from roadway sensors including thermometers and moisture detectors — i.e., weather-derived roadway conditions.
2. If the effective filing date is 2015-09-25, US 9,785,145 is only §102(a)(2) art (it published after that date), and because the '446 and the '145 were both owned by / subject to assignment to IBM as of 2015-09-25 (the inventor declarations and assignments in the prosecution records confirm IBM as assignee), §102(b)(2)(C) likely removes it from the prior art entirely — including for §103. This is my leading hypothesis for why the weather limitation was added and why the claims issued over it.

I flag one further fact without over-reading it: the reference listing printed on the '446 does not appear to include US 9,785,145 or US 2017/0038773 A1, even though it includes IBM siblings such as US 9,483,948, US 9,513,632 and US 9,566,986 and lists a "List of IBM Patents or Patent Applications Treated as Related." Because that listing is OCR-derived, I cannot state this with confidence, and I have not pulled the file wrapper.

---

3. Reference-by-reference mapping (claim 1)

Correction to avoid mischaracterization: US 8,078,349 B1 is "Transitioning a mixed-mode vehicle to autonomous mode" (Google) — a landing-strip/reference-indicator handoff, not a driver-skill-comparison reference. I verified its actual content (Google Patents US8078349B1, PDF). It is good evidence that automated/manual mode switching was known, but it should not be cited as teaching a competence comparison.

---

4. Motivating combinations

Combination A (strongest, if the reference is available): US 9,785,145 alone or + any weather-condition mode-control reference

• What each teaches: '145 teaches every step of claim 1 except the express "weather" cause of the roadway condition. Google/Waymo zone-driving art (US 8,660,734; US 2014/0324268) and Google's Engaging and disengaging family teach that permission to run autonomously depends on the environment/roadway conditions about the vehicle.
• Motivation: (i) '145's own specification already enumerates "weather conditions" among the roadway conditions it monitors and names a weather station as a data source — so the substitution is disclosed, not merely suggested; (ii) adverse weather (rain, snow, ice, fog, glare) was the paradigmatic reason in 2015 to question whether either a perception-limited automated system or an inattentive human should be in control; (iii) combining two condition-based mode-control teachings in the same field, addressing the same problem, with no change in principle of operation, is a textbook KSR combination (known elements, predictable result).
• Reasonable expectation of success: high — the combination only adds a sensor input (weather) to an already-disclosed comparison, with no new structural or algorithmic requirement.

Combination B (available regardless of priority): mode-arbitration art + driver-scoring art + the ip.com publications

• Mode arbitration: US 9,317,033 (bimodal autonomous/manual control), US 2015/0057891 (context-aware arbitration of control authority between driver and automation), US 9,381,915/US 9,075,413 (environmental + system + driver checks before engaging/disengaging automation).
• Driver competence: the cited usage-based/telematics insurance references (US 8,140,358; US 8,090,598; US 7,877,269; US 9,390,451) establish, as of well before 2015, the routine derivation of driver-specific skill/risk profiles from historical driving data.
• System competence: the cited ip.com item No. 000238474 (validating self-driving capability) and ip.com No. 000234916 (built-in mechanism to enforce mandatory self-driving mode).
• Motivation: each reference addresses the same problem — "who should be driving right now?" — and the combination is a mere aggregation of known, independently conventional data sources (a driver risk profile and a system capability/health assessment) fed into a known arbitration decision. Where the prior art teaches that both candidate operators have condition-dependent competence, ranking them and assigning control to the better one is the predictable use of prior-art elements according to their established functions (KSR; MPEP 2143.01(IV)). The ip.com publications independently show the field's recognition that autonomy should sometimes be mandated — supplying the "policy" reason to make the comparison and act on it.

Combination C — evidence within the patent family itself (highly probative, with a caveat)

The family's independent claims differ almost exclusively in the trigger used for the same CPCL/HDCL comparison: US 9,785,145 / US 10,029,701 (roadway condition / operational anomaly), US 9,566,986 (operational anomaly), the '446 (weather), and the '446's own claim 20 (lighting). Treating the trigger as a mere substitution of one known environmental condition for another is a routine design choice where the art already recognizes that different conditions differentially impair human and machine drivers. Caveat: the family members are co-owned and largely §102(b)(2)(C)-excepted, so this is best used as evidence of what the art considered conventional rather than as a §102 reference.

---

5. Dependent claims

---

6. Where the §103 case is weak (counter-arguments the patent owner will press)

1. The weather clause may be the whole case. If US 9,785,145 is §102(b)(2)(C)-excepted, the examiner had to find the comparison conditioned on weather elsewhere. None of the references I can verify with confidence discloses the pair of steps — "quantify the automated system's competence level" and "quantify the human's competence level" both conditioned on a weather-derived road condition — and then rank them. A patent owner will argue the asserted combinations are hindsight reconstructions built from the claims themselves.
2. "Competence level" is a term of degree. No reference I verified teaches computing a scalar competence metric for a control processor at all; the closest art assesses capability, health, or availability. Expect an argument that the claims are not enabled/definite—but, more to the point here, that the art does not teach a comparison of two such metrics.
3. Two-directional assignment. Claim 1 covers assignment to the human as well as to the processor. Several of the strongest references (e.g., the ip.com "mandatory self-driving mode" item, zone-driving art) only teach prohibiting autonomous operation, i.e., pushing toward manual — the opposite direction. A combination would need both directions to be motivated.
4. No secondary considerations of record. Priority/sale of the portfolio to Daedalus → Slingshot IoT → Granite Vehicle Ventures indicates the patents were asserted/valued, but there is no evidence of nexus, industry praise, licensing-due-to-merits, or unexpected results that I have found. Absence of objective indicia does not defeat a prima facie case, but it removes a powerful rebuttal.

---

7. Stated uncertainties and verification limits

• Priority determination unresolved. I have not reviewed 14/865,393 line-by-line to confirm §112 support for the "result of current weather conditions" clause. Everything turns on this.
• Reference texts not all retrieved. US 9,317,033, US 2015/0057891, US 9,399,472, US 8,140,358, US 8,090,598, US 7,877,269, US 9,390,451, US 9,024,787, US 9,189,897, and the WO 2014/147361, WO 2014/148975, WO 2014/148976, WO 2015/024616, WO 2015/056105, WO 2014/058263 documents are cited on the '446 face but I did not retrieve their text. Titles/assignees for four of them come from the Unified Patents art list, not from the documents themselves; treat them as unverified.
• US 9,381,915 characterization. I inferred its subject matter from the Google "Engaging and disengaging for autonomous driving" family (US 9,075,413 / US 2017/0253253). I did not retrieve US 9,381,915's own text. Verify before relying on it.
• US 2016/0355192 A1. The '446 face lists it as "James et al." (Dec. 2016); a cross-listing for sibling US 10,109,195 gives US-20160355192-A1 as Toyota, "Transitioning Between Operational Modes of an Autonomous Vehicle," priority 2015-06-03. If that Toyota characterization is correct, it is a strong, non-commonly-owned §102(a)(2) reference (priority — and thus its art date — is before both candidate effective filing dates). I could not reconcile the two listings; flagging rather than resolving.
• File wrapper not reviewed. I have not seen the examiner's reasons for allowance or the IDS. My hypothesis that US 9,785,145 was the near-miss reference is an inference, not a finding.
• Consistency with prior sections. The earlier sections' negative litigation finding for the '446 stands; nothing here concerns the '402/'765/'004 patents asserted in Granite Vehicle Ventures v. Tesla. The earlier-noted legal-status conflict ("Expired – Fee Related" vs. "In Force") does not affect a §103 analysis.

---

8. Bottom line

• Most likely invalid under §103 if the '446 is not entitled to the 2015-09-25 priority date: US 9,785,145 / US 2017/0038773 A1 discloses claim 1 essentially verbatim minus the weather clause, and its own specification supplies weather-derived roadway conditions. Claims 2–9 and 20 are then obvious over that reference plus the cited driver-scoring, zone-driving, ip.com, and geometry references.
• If the priority date holds at 2015-09-25, US 9,785,145 is probably removed from the art by §102(b)(2)(C), and the §103 case becomes a genuine two-reference combination (mode-arbitration art such as US 9,317,033 / US 2015/0057891 / US 9,381,915 + driver-scoring/telematics art, with the cited ip.com publications supplying the "mandate autonomy" policy). That combination is reasoned and defensible — the motivation is strong (same field, same problem, known elements, predictable result per KSR and MPEP 2143.01) — but it is not a certainty, because the record I can verify does not show a single reference expressly ranking a control-processor competence metric against a human competence metric under a weather-derived roadway condition.
• Practical recommendation: the highest-value next step is not more art searching but (a) confirming the §112 support for the weather clause in 14/865,393, and (b) obtaining the '446 file wrapper, since the IDS contents and the examiner's stated reasons for allowance will reveal whether the art relied on here was already before the Office.