Patents in figures

1. How many: annual US patent grants

IFI Claims ranking: patents granted to Apple each year. Method limit: IFI states on its 2016 pages that it counts only utility patents; this has not been confirmed for 2024-25, so those two figures may not be comparable. 2014 and the years before 2013 are missing.

Sources: IFI Claims, Top 50 tables for each year (https://www.ificlaims.com/rankings/top-50-2023/ and equivalents 2013-2022); 2024 per IFI table contributed to USPTO (ptacts.uspto.gov); 2025 by IFI note and MacRumors (https://www.macrumors.com/2026/01/13/apple-patents-declined-in-2025/).

Table IFI2024 incorporated into the USPTO file · IFI method

2. What a patent looks like: examples with drawings

A document identifies number and type, title, inventors, assignee and dates. Includes drawings, description and claims. In a design patent, the figures are central. It is not a guide to processing a patent.

Text and figures from granted patents. The USPTO says text and drawings are normally free of copyright restrictions unless a notice states otherwise (37 CFR 1.71(d) and 1.84(s)); no such notice was found in these five examples. This is not legal advice. USPTO Rule and Exceptions.

3. Patent and product: documented links only

Unit = commercial product. Several patents for the same iPhone are not independent cases. A generic title does not prove or rule out a link; The criterion is document plus drawing.

Sources: https://www.cafc.uscourts.gov/opinions-orders/14-1335.opinion.5-14-2015.1.pdf · https://patents.justia.com/patent/D756823 · https://patents.justia.com/patent/D616892 · https://www.invent.org/inductees/steve-wozniak · https://www.cafc.uscourts.gov/opinions-orders/12-1105.pdf · https://www.mactech.com/2010/06/01/apple-granted-design-patents-for-the-ipad-ipods-macbook-air-more/amp/

4. Documented release schedule

There are five products from different periods: they serve to describe, not to set a typical deadline.

Sources: Apple newsroom https://www.apple.com/newsroom/2001/10/23Apple-Presents-iPod/ · https://www.apple.com/newsroom/2001/11/09Apple-s-iPod-Available-in-Stores-Tomorrow/ · https://www.apple.com/newsroom/2007/01/09Apple-Reinvents-the-Phone-with-iPhone/ · https://www.apple.com/newsroom/2007/06/28iPhone-Premieres-This-Friday-Night-at-Apple-Retail-Stores/ · https://www.apple.com/newsroom/2010/01/27Apple-Launches-iPad/ · https://www.apple.com/newsroom/2010/03/05iPad-Available-in-US-on-April-3/ · https://www.apple.com/newsroom/2014/09/09Apple-Unveils-Apple-Watch-Apples-Most-Personal-Device-Ever/ · https://www.apple.com/jp/newsroom/2015/03/09Apple-Watch-Available-in-Nine-Countries-on-April-24/ · Apple II: https://www.cultofmac.com/apple-history/apple-ii-launch and https://www.computerhistory.org/tdih/June/10/

5. What becomes public, and when

• In the US, ordinary design applications are not subject to the 18-month publication rule; the document becomes public on grant. Utility patent applications are usually published about 18 months from the earliest recognized date, with exceptions such as non-publication requests or secrecy orders. Provisional applications are not published under that rule. Their contents may still become public in a later case record.
• In these cases, little information was public before launch. The iPhone design patents were granted after it went on sale (D593,087: filed 2007-07-30, a month later). Exception: iPad design D504,889 was public almost five years before the announcement, but its product link is only an allegation by Apple.
• This describes this sample. It supports neither "almost always after launch" beyond this sample nor the opposite.

USPTO: publication and exceptions

6. The 50 designs in a court record

Appendix to Apple v. Samsung (document 3647-5, 2018-04-09): 50 distinct designs asserted by Apple. They were selected for litigation, not a census or a representative sample of Apple, and the appendix does not identify a product for each design. They are an example of a dated dataset.

Source: https://storage.courtlistener.com/recap/gov.uscourts.cand.239768/gov.uscourts.cand.239768.3647.5.pdf (number, filing and grant dates for the 50 rows extracted from the PDF). Subtotals by title type have not been cleaned and are not published.

Download the 50 rows of the annex (CSV)

7. Secrecy: what is known and what is not

It is treated as an observable dimension, without access to internal plans and without sentiment score. What is documented in the press and books:

• The New York Times (2009) describes Apple's secrecy as increasing, with firings over leaks and misinformation about product plans: https://www.nytimes.com/2009/06/23/technology/23apple.html
• Regarding the iPhone (codename Purple), CACM and The Verge report a small team, of people already at Apple, working behind closed doors: https://cacm.acm.org/news/the-secretive-life-of-apple/ and https://www.theverge.com/2017/6/13/15782200/one-device-secret-history-iphone-brian-merchant-book-excerpt
• Ars Technica (2012) says that, despite everything, the products have always been leaked: https://arstechnica.com/gadgets/2012/10/apples-secret-garden-the-struggle-over-leaks-and-security/

8. Hypotheses about lead times (to be tested)

The documented historical dates and links are facts, and so are the gaps measured with them. It is a hypothesis to extrapolate them to future products; They can be used for scenarios, always labeled.

Launch forecasts: experimental design

A forecast is a reasoned, traceable hypothesis, not a certainty. No system has been implemented and no probabilities calibrated; calibration would require measuring outcomes first.

1. Each forecast is written before the fact, with the date: product or family, expected window, and the public evidence that supports it (published applications, granted designs, Apple's previous calendar).
2. Record what would refute it (for example, no announcement within the forecast window) and the uncertainty level in words: low, medium or high.
3. It is saved in a record that is not edited. When the date arrives, it is compared with the real date in the Apple newsroom.
4. Only after enough forecasts have reached their outcome can we test whether they beat a simple baseline (frequency per quarter). Until then, they are hypotheses.

9. Still to check

• Status (expired, abandoned, maintenance fees unpaid): not yet checked in the relevant records. It is not inferred from a patent's age.
• Latest published applications: coverage still limited, no systematic review.
• Complete annual series (2014 and before 2013) and own total: pending; The official bulk download consulted requires an account.
• Links for iPod, Watch, Mac, AirPods, HomePod and Apple TV.