PTAB

IPR2020-01611

Google LLC v. Human Differential Intelligence LLC

Log In
Key Events
Petition
petition

1. Case Identification

2. Patent Overview

  • Title: Completely Automatic Public Turing Test to Tell Computers and Humans Apart (CAPTCHA)
  • Brief Description: The ’861 patent describes methods for authenticating a human user by issuing a CAPTCHA challenge that requires physical interaction with a portable computing device. The challenges involve either manipulating the device itself (e.g., rotating or shaking it) or interacting with its touchscreen (e.g., performing multi-touch gestures).

3. Grounds for Unpatentability

Ground 1: Anticipation/Obviousness over Seacat - Claims 12 and 16 are unpatentable over Seacat under 35 U.S.C. §102/§103.

  • Prior Art Relied Upon: Seacat (Patent 8,214,891).
  • Core Argument for this Ground:
    • Prior Art Mapping: Petitioner argued that Seacat, which was not substantively considered during prosecution, teaches all limitations of the touch-based CAPTCHA claims. Seacat discloses using a multi-touch portable device to present a CAPTCHA challenge, such as instructing a user to rotate an on-screen image of a star by 90 degrees using their fingers. The system automatically elects the challenge, processes the user’s touch sensory data, and determines if the input meets a required threshold (e.g., rotation between 85-95 degrees) to qualify the user as human. This directly maps to the limitations of independent claim 12. Dependent claim 16, which requires the challenge to be embodied in computer-readable form, was allegedly met by Seacat’s disclosure of a “client application” implementing the CAPTCHA test.
    • Motivation to Combine (for §103 grounds): N/A for anticipation. For obviousness, Petitioner asserted that to the extent Seacat does not explicitly disclose that its steps are "automatic" or use a "threshold," implementing these features would have been a conventional and obvious design choice for a POSITA creating a CAPTCHA system.
    • Expectation of Success (for §103 grounds): A POSITA would have had a high expectation of success in implementing these conventional features, as they were standard in the art of automated security challenges.

Ground 2: Obviousness over Seacat and Mattice - Claims 1, 5, and 9 are obvious over Seacat in view of Mattice.

  • Prior Art Relied Upon: Seacat (Patent 8,214,891) and Mattice (Application # 2007/0259716).
  • Core Argument for this Ground:
    • Prior Art Mapping: This ground addresses the "motion and/or orientation-based" claims under a narrower interpretation requiring movement of the device itself. Petitioner argued that Seacat provides the foundational automated CAPTCHA method involving user manipulation of on-screen objects. Mattice, also not considered during prosecution, teaches using a portable device’s motion and orientation sensors (e.g., accelerometers) as a user interface for tasks including authentication and security. The combination allegedly renders claim 1 obvious by applying Mattice's motion-based user interface to Seacat's CAPTCHA framework. For example, a user could satisfy Seacat's "rotate the star" challenge by physically rotating the device itself, as enabled by Mattice's teachings.
    • Motivation to Combine: A POSITA would combine these references to improve the CAPTCHA system. The combination would offer greater flexibility (providing an alternative to touch input for users with motor impairments) and enhanced security (allowing for multi-modal challenges, e.g., both touch and motion). Combining known user interfaces with security applications was a predictable variation.
    • Expectation of Success: A POSITA would have reasonably expected success, as it involved combining two known technologies—a motion-based user interface and a CAPTCHA system—for their intended and understood purposes.

Ground 3: Obviousness over Seacat and Shirali-Shahreza - Claims 4 and 14 are obvious over Seacat in view of Shirali-Shahreza.

  • Prior Art Relied Upon: Seacat (Patent 8,214,891) and Shirali-Shahreza (a 2008 IEEE conference paper).

  • Core Argument for this Ground:

    • Prior Art Mapping: This ground challenges dependent claims 4 and 14, which add the limitation of presenting instructions in "audible form by a text-to-speech engine." Petitioner contended that Seacat provides the underlying touch-based and motion-based CAPTCHA challenges. Shirali-Shahreza, not considered during prosecution, explicitly teaches making CAPTCHA challenges more accessible to children or the visually impaired by presenting instructions audibly using a Text-To-Speech (TTS) engine on portable devices like PDAs or mobile phones.
    • Motivation to Combine: A POSITA would be motivated to add Shirali-Shahreza's TTS functionality to Seacat's CAPTCHA system to improve accessibility and broaden the range of potential users, a well-known goal in user interface design. Both references address CAPTCHA on portable devices, making the combination a natural and predictable improvement.
    • Expectation of Success: The combination was a simple application of a known technique (TTS for instructions) to an existing system (Seacat's CAPTCHA) to achieve a predictable result (an audible CAPTCHA challenge). The technical implementation would have been straightforward for a POSITA.

  • Additional Grounds: Petitioner asserted that claims 1, 5, and 9 are anticipated or obvious over Seacat alone under the Patent Owner's broad claim construction. It also asserted an additional obviousness challenge for claim 4 based on the combination of Seacat, Mattice, and Shirali-Shahreza.

4. Key Claim Construction Positions

  • Petitioner advanced arguments under two alternative constructions for claim 1's limitation "motion and/or orientation based... physical manipulation instructions."
    • Broad Interpretation: Aligning with the Patent Owner's alleged litigation position, this covers any physical movement by the user, including hand or finger movements to interact with a GUI (e.g., rotating an on-screen object). Under this view, Seacat alone was argued to anticipate claim 1.
    • Narrow Interpretation: This construction requires physical movement or re-orientation of the portable computing device itself. Petitioner argued that even under this narrower reading, the claims are obvious over Seacat in view of prior art teaching motion-based user interfaces, such as Mattice.

5. Key Technical Contentions (Beyond Claim Construction)

  • Petitioner contended that none of the Challenged Claims are entitled to the filing date of the ’861 patent’s provisional application. It argued the provisional, which focused on an electronic Pictionary-style game, completely lacks written description support for key terms recited in the claims, such as "motion and/or orientation," "touch-based," "sensory data," and "physical manipulation instructions." Therefore, the effective priority date is no earlier than the non-provisional filing in June 2009, making references like Seacat, Mattice, and Shirali-Shahreza prior art under any legal framework.

6. Arguments Regarding Discretionary Denial

  • Petitioner argued against discretionary denial under §314(a) (Fintiv factors), asserting that: (1) Google is not a defendant in the parallel district court litigations; (2) the litigations were in very early stages with minimal investment and no substantive orders issued; (3) the trial date was speculative and likely to slip; and (4) the strong merits of the petition, based on art the examiner never considered, weigh heavily in favor of institution.
  • Petitioner also argued against denial under §325(d), contending that the Examiner made a material error by failing to appreciate the teachings of the key prior art. Although Seacat's published application was listed in a 17-page IDS with over 150 references, it was never cited, discussed, or applied in any rejection. Mattice and Shirali-Shahreza were never before the Examiner at all.

7. Relief Requested

  • Petitioner requested institution of an inter partes review and cancellation of claims 1, 4-5, 9, 12, 14, and 16 of the ’861 patent as unpatentable.