Background
Wildseed Mobile owns U.S. Patent No. 7,376,414, titled “Method and System for Inserting Advertisements into Broadcast Content.” The patent describes a system for delivering targeted advertisements to mobile device users by collecting information about them — including from text messages sent by other users — and using that data to select relevant ads. When the system determines it is time for an ad, it pauses the user’s audio content, plays the ad, records the missed content, and resumes playback afterward.
Google petitioned the Patent Trial and Appeal Board (PTAB) for inter partes review (IPR), arguing that claims 1–7 and 9–14 of the patent were obvious over a combination of three prior art references: Frerichs (a system for inserting ads into streamed audio based on user profiles), Kawasaki (a method for generating user interest profiles from messages), and Kauffman (seamless data insertion for streaming media).
The central dispute at the Board came down to claim construction. The key claim language required the system to collect “information taken by the cellular device from a message sent to the cellular device from another user.” Wildseed argued this was limited to extracting the exact words of a text message — so if someone texts “Let’s play Pebble Beach today,” the system could only pass along those literal words, not deduce that the user is interested in golf. Google argued the term includes information derived or generated from the message, such as inferring a golf interest from the mention of a famous golf course. The Board sided with Google, and on that basis found all challenged claims obvious.
The Court’s Holding
The Federal Circuit affirmed, agreeing with the Board on both disputed claim constructions. Judge Stark, writing for a unanimous panel, held that nothing in the claim language, specification, or prosecution history limits “taken from a message” to verbatim extraction. The court found the claim language specifies the source of the information (a text message) but does not constrain how that information is processed or used once obtained.
The court rejected Wildseed’s argument that the doctrine of claim differentiation supported its narrow reading. Under the Board’s construction, “taken” encompasses both verbatim extraction and derived information, while “determined” in other claim terms refers only to generated information — giving each term distinct scope.
Wildseed also argued that the Board’s construction improperly applied hindsight by assuming capabilities beyond 2001-era cellphones. The court disagreed, noting that even early-2000s technology could perform database lookups and simple matching — for instance, associating “Pebble Beach” with golf courses — without requiring modern AI or neural networks.
On the dependent claims, which add categories like user preferences, demographics, or location to the information sent to the server, the court similarly agreed with the Board that these additional information types need not come exclusively from text messages. Finally, the court found no APA violation in the Board’s analysis, holding that its reasoning was sufficiently clear to enable judicial review.
Key Takeaways
- Broad construction of “taken from” language: When a patent claims information “taken from” a particular source, that language identifies where the information comes from, not whether it can be processed, derived, or interpreted along the way. Patent drafters seeking narrow extraction-only scope must use explicit limiting language.
- Prosecution history cuts both ways: Wildseed’s amendment during prosecution narrowed the source of information (to text messages from other users) to overcome prior art. But the court held this narrowing said nothing about the method of obtaining information from that source, undermining Wildseed’s argument that derived information was disclaimed.
- Hindsight objections require specificity: Arguing that prior art combinations would not have been obvious to a person of ordinary skill at the time of invention requires showing specific technical limitations, not general assertions about the state of technology.
Why It Matters
This decision illustrates the risks of relying on vague claim language to distinguish prior art, especially in the fast-moving digital advertising space. Wildseed’s patent described a system for targeted mobile advertising — a concept that quickly became ubiquitous — and the claims were not drafted tightly enough to survive an obviousness challenge combining three references that collectively taught the same building blocks.
For patent holders in the ad-tech and mobile technology spaces, the case underscores the importance of precise claim drafting. Terms like “taken from” will be read broadly unless the specification or prosecution history clearly anchors them to a specific technical implementation. For challengers, the decision confirms that the PTAB’s broad claim construction approach continues to enjoy Federal Circuit support, making IPR an effective tool for clearing weak patent claims from the field.
Your browser cannot display this PDF inline.
Download the full opinion (PDF)