Table of Contents

A patent infringement analysis San Francisco litigators can rely on has to keep pace with a startup economy where the accused product is often an AI model, a payments API or a biologic rather than a finished machine. San Francisco anchors a distinct slice of the innovation economy — Salesforce, Uber, Airbnb, Dropbox, Pinterest and a wave of AI companies including OpenAI, Anthropic and Databricks are based in the city, alongside a deep fintech and life-sciences base that runs down the Peninsula.
A company accused of infringing a software, AI, fintech or biotech patent here is typically sued in the San Francisco Division of the U.S. District Court for the Northern District of California, or it challenges the patent at the PTAB and defends imports at the ITC. In each forum the case turns on a disciplined element-by-element comparison of the accused product against the asserted claims. PerspireIP builds that infringement and evidence-of-use analysis for the companies asserting and defending patents across San Francisco.
Where a patent infringement analysis San Francisco case is heard
Patent suits arising in the city are heard in the U.S. District Court for the Northern District of California, and cases from San Francisco, Marin and the surrounding counties are assigned to its San Francisco Division. The division sits in the Phillip Burton Federal Building and United States Courthouse at 450 Golden Gate Avenue, in the Civic Center. Patent validity and infringement are exclusively federal questions — there is no state-court patent venue in California, and everything from claim construction to a jury verdict runs before a federal judge.
Under the Supreme Court’s decision in TC Heartland, a company can be sued for patent infringement only where it is incorporated or where it has a regular, established place of business and has committed acts of infringement. Because so many technology and life-sciences companies are incorporated in Delaware but headquartered in San Francisco, venue in the San Francisco Division is a live, heavily litigated question, and the presence of a local headquarters, lab or engineering office often decides it. Appeals do not follow the usual path: while most Northern District appeals go to the Ninth Circuit, every patent appeal goes instead to the Court of Appeals for the Federal Circuit in Washington, D.C.
- N.D. Cal., San Francisco Division — the Phillip Burton Courthouse, where infringement and validity are tried to the court and to a jury
- PTAB — inter partes review, decided nationally on novelty and obviousness over patents and printed publications
- ITC — Section 337 exclusion actions for imported devices and goods, where infringement and invalidity are both in play
- Federal Circuit — all patent appeals, from the district court, the PTAB and the ITC
The San Francisco Division draws a different mix of disputes from the semiconductor-heavy San Jose Division down the road — more software, AI, fintech and life-sciences assertions — but the same Patent Local Rules and the same demanding, technically fluent bench. The party that arrives with a rigorous, well-charted infringement or non-infringement position, rather than a conclusory theory, controls the narrative from the first case management conference.
San Francisco’s AI, fintech, software and biotech patents
San Francisco’s inventive record leans toward software, platforms, financial technology and the life sciences rather than the chips and hardware that define the South Bay. Salesforce, Uber, Airbnb, Dropbox, Pinterest, Twilio and Block are based in the city, a dense fintech cluster runs from Stripe to established payments and lending platforms, and the current wave of generative-AI companies — among them OpenAI, Anthropic and Databricks — has made the city a global center for machine-learning innovation. Genentech and a broad biotech corridor extend the base south toward South San Francisco.
That mix shapes exactly where a patent infringement analysis San Francisco case has to look for evidence. The asserted claims are frequently method and system claims covering software, distributed architectures, machine-learning models, payment flows and user-interface techniques — subject matter where infringement is proven from behavior and code, not from a physical teardown. Life-sciences and diagnostics patents add a second, very different track built on assays, sequences and clinical data.
- Software, cloud and platform method and system claims, including SaaS and API architectures
- Artificial-intelligence and machine-learning claims covering model training, inference and data pipelines
- Fintech and payments claims — transaction processing, fraud detection and digital-wallet techniques
- Biotech, diagnostics and medical-device claims tied to the region’s life-sciences base
Because so many of these claims are software and business-method inventions, Alice eligibility under Section 101 is a constant companion to the infringement question. A defensible analysis reads the claim for both what it covers and whether it is directed to an abstract idea, so counsel can weigh infringement, validity and eligibility together.
N.D. Cal. Patent Local Rules and the contentions calendar
The Northern District of California was the first court in the country to adopt Patent Local Rules, back in 2000, and they govern the San Francisco Division exactly as they govern San Jose. The rules force both sides to commit to specific positions early, long before the close of discovery, which is precisely why front-loaded infringement and non-infringement analysis pays off in this district. A vague theory that would survive elsewhere gets exposed here on a fixed schedule.
The sequence is unforgiving. Under Patent Local Rule 3-1, the patentee must serve Disclosure of Asserted Claims and Infringement Contentions — a claim chart mapping every asserted claim element to the accused product or service — within 14 days of the initial Case Management Conference, along with the document production required by Rule 3-2. The accused infringer then serves Invalidity Contentions under Rule 3-3 within 45 days. The 2017 revisions added early damages contentions on top of that, tightening the timeline further. Claim construction under Patent Local Rules 4-1 through 4-5 follows, culminating in a Markman hearing.
- Rule 3-1 / 3-2 — the patentee’s Infringement Contentions and document production, due 14 days after the Case Management Conference
- Rule 3-3 / 3-4 — the accused infringer’s Invalidity Contentions and document production, due 45 days later
- Rules 4-1 to 4-5 — the claim-construction exchange, joint statement, briefing and Markman hearing
For a software or AI patent the early chart is especially challenging, because the accused behavior often lives in a service the patentee cannot fully see before discovery. The patentee that pleads its Rule 3-1 contentions on the best available public evidence — then refines them from produced source code — keeps its theory alive; the accused party that documents the gaps early builds its non-infringement case around them.
Evidence of use and claim charts for software and AI claims
Infringement is proven element by element, and in San Francisco the accused element usually lives inside a cloud service, a mobile app or a trained model that no one can inspect from the outside. Building evidence of use (EoU) means capturing the actual behavior of the product, not its marketing. For a software or fintech claim that can mean API traces, network-traffic captures, app decompilation and, in litigation, produced source code; for an AI claim it can mean documenting model architecture, training data and inference behavior. A serious analysis reaches those sources directly.
- API documentation, network-traffic captures and app decompilation for software, platform and fintech claims
- Model cards, technical papers and observed inference behavior for machine-learning and AI claims
- Source-code review under a protective order, the gold standard for software and AI method claims
- Assay protocols, sequence data and clinical disclosures for biotech and diagnostics claims
- Datasheets, white papers and standards specifications for public-facing proof
The output is a claim chart a judge and jury can follow: each limitation in its own row, the corresponding step or component in the accused product beside it, and a citation to the evidence that proves it. We build charts to the standard the San Francisco Division expects — specific, sourced and defensible — whether they support a patentee’s Rule 3-1 contentions or an accused party’s non-infringement position. A chart that hand-waves past a single limitation is a chart that loses on summary judgment.
IPR at the PTAB, district court, or the ITC
A San Francisco litigant usually has more than one forum available, and they are not interchangeable. For an accused infringer, inter partes review at the PTAB is a fast, cost-effective way to attack validity — but the grounds are limited to novelty and obviousness over patents and printed publications, and IPR does not resolve infringement at all. Infringement is decided only in the district court or at the ITC, which is why an accused party almost always needs a non-infringement analysis running in parallel with any IPR.
District-court litigation in the San Francisco Division decides the whole dispute — infringement, validity, Section 101 eligibility, damages and, where available, willfulness and injunctive relief. It is broader and slower than an IPR, and it applies the higher clear-and-convincing standard to invalidity. The Northern District’s Patent Local Rules and its judges’ fluency with software and life-sciences technology make it a forum where a well-built infringement or non-infringement record genuinely moves the case, at claim construction and again at summary judgment.
Timing drives the choice. A defendant served with a complaint must file its IPR within one year, and an IPR that reaches a final written decision carries estoppel on grounds raised or that reasonably could have been raised. A patentee weighing the ITC gains speed and an exclusion-order remedy but takes on the domestic-industry requirement. Many San Francisco disputes therefore run on parallel tracks, with one rigorous infringement and evidence-of-use record built to feed all of them.
The ITC, Section 337 and imported devices
Even a software-led economy runs on imported hardware — phones, wearables, medical devices and connected products — and that pulls some San Francisco disputes toward the U.S. International Trade Commission. Under Section 337, the ITC investigates imported goods accused of infringement and can issue an exclusion order barring them from entry — a fast, powerful remedy that can run alongside, or instead of, a district-court suit. For a company that sells an imported device or connected product, that threat is serious.
At the Commission, infringement is the central question, and it moves on the ITC’s compressed timeline before an administrative law judge. A patentee must also prove a domestic industry — U.S. investment in articles practicing the patent, which can include substantial engineering and R&D — while a respondent builds non-infringement and invalidity in parallel. The technical mapping of the accused import against the asserted claims has to be located, charted and evidenced before the schedule closes, which arrives far sooner than in district court. A company that starts its infringement analysis at the first sign of a complaint keeps every option open.
How PerspireIP builds a patent infringement analysis San Francisco case
Every engagement follows the same disciplined path. We construe the asserted claims, break each one into its limitations, and then investigate the accused product or service against those limitations one by one — using API and traffic analysis, app decompilation, model and architecture documentation, source-code review and, for life-sciences claims, assay and sequence analysis as the technology requires. We build evidence-of-use and claim charts a San Francisco Division judge, a PTAB panel or an ITC administrative law judge can follow, sized to the deadline that actually governs your case.
- Element-by-element claim mapping for infringement or non-infringement, tied to a defensible claim construction
- Software and fintech evidence of use — API traces, network captures, decompilation and produced-source-code review under protective order
- AI evidence of use — model architecture, training-data and inference-behavior analysis for machine-learning claims
- Section 101 eligibility read alongside infringement for software and business-method claims
- Charts sized to your forum’s clock — the Patent Local Rule 3-1 deadline, the PTAB’s one-year bar or the ITC’s fast track
- A written analysis and exhibit-ready charts for the district court, the PTAB or the Commission
We work alongside your California litigators and patent counsel as a specialist technical partner, deliver to Northern District, PTAB and ITC deadlines, and keep every engagement confidential. Whether you are a patentee preparing Rule 3-1 infringement contentions, an accused startup building a non-infringement and invalidity defense, or an importer facing a Section 337 complaint, we scale to fit — a single product read, a multi-patent campaign or ongoing support. Send us the patent number and the accused product, and we will scope a patent infringement analysis San Francisco project within one business day.
IP Landscape & Resources in San Francisco
Key intellectual-property authorities and venues relevant to San Francisco:
- U.S. District Court for the Northern District of California — the federal trial court whose San Francisco Division (Phillip Burton Courthouse, 450 Golden Gate Ave) hears the city's patent infringement and validity cases
- N.D. Cal. Patent Local Rules — the district's dedicated patent rules setting the infringement-contentions, invalidity-contentions and claim-construction calendar
- USPTO Patent Trial and Appeal Board (PTAB) — administers inter partes review, the administrative route to challenge patent validity on prior art
- U.S. International Trade Commission โ Section 337 — investigates imported devices and connected products accused of infringement and can issue exclusion orders
Request a Patent Infringement Analysis in San Francisco
Request a Patent Infringement Analysis in San Francisco
Get an element-by-element infringement or non-infringement analysis built for the Northern District of California’s San Francisco Division, the PTAB and the ITC, with evidence of use and claim charts tuned for AI, software, fintech and biotech claims. Send us the patent number and the accused product, and we will scope the work within one business day.
Explore related PerspireIP services: Patent Infringement Analysis · Patent Invalidation · Prior Art Litigation Search.
Frequently Asked Questions
Which court hears San Francisco patent cases?
Patent suits are exclusively federal. Cases arising in the city are filed in the U.S. District Court for the Northern District of California, and matters from San Francisco and the surrounding counties are assigned to its San Francisco Division in the Phillip Burton Federal Building at 450 Golden Gate Avenue. Patent appeals do not go to the Ninth Circuit; they go to the Court of Appeals for the Federal Circuit in Washington, D.C. Validity can also be challenged nationally at the USPTO’s PTAB through inter partes review, and for imported goods at the International Trade Commission under Section 337.
Do the N.D. Cal. Patent Local Rules apply in San Francisco?
Yes. The Northern District of California was the first court to adopt Patent Local Rules, in 2000, and they govern the San Francisco Division exactly as they govern San Jose. Under Patent Local Rule 3-1 the patentee must serve Infringement Contentions โ a claim chart mapping each asserted element to the accused product or service โ within 14 days of the Case Management Conference, and the accused infringer serves Invalidity Contentions under Rule 3-3 within 45 days. Claim construction and a Markman hearing follow. The early, fixed calendar rewards parties that arrive with rigorous infringement or non-infringement analysis already in hand.
How is infringement of an AI or software patent proven in San Francisco?
From behavior and code, not from a physical teardown. For a cloud, fintech or AI service the evidence of use is built from API traces, network-traffic captures, app decompilation, model and architecture documentation, and โ in litigation โ source code produced under a protective order. Because many of these are software and business-method claims, Section 101 eligibility under Alice is analyzed alongside infringement. We chart each claim limitation against the accused service from the strongest available evidence so the analysis holds up at claim construction and on summary judgment in the San Francisco Division.
Should an accused startup in San Francisco file an IPR or defend in district court?
Often both. Inter partes review at the PTAB is a fast, lower-cost way to attack validity, but it is limited to novelty and obviousness over patents and printed publications and does not decide infringement. Infringement โ and Section 101 eligibility โ are resolved only in the district court or at the ITC, so an accused party almost always needs a non-infringement analysis running alongside any IPR. A defendant served with a complaint has one year to file an IPR, and a final written decision carries estoppel, so the forum decision and the supporting analysis should be made early.