Google LLC v. Security First Innovations LLC

1. Case Identification

• Case #: IPR2024-00213
• Patent #: 9,338,140
• Filed: November 27, 2023
• Petitioner(s): Google LLC
• Patent Owner(s): Security First Innovations, LLC
• Challenged Claims: 1-10

2. Patent Overview

• Title: Data Storage System and Method
• Brief Description: The ’140 patent discloses a data storage system that enhances security by splitting a dataset into multiple "shares," encrypting the data using a key, and distributing the shares to different storage locations. A central feature is storing the encrypted data shares along with data indicative of the key used to secure the dataset.

3. Grounds for Unpatentability

Ground 1: Obviousness over Ophir and Takeda - Claims 1-6, 8, and 10 are obvious over Ophir in view of Takeda.

• Prior Art Relied Upon: Ophir (Application # 2004/0143733) and Takeda (Application # 2001/0053221).
• Core Argument for this Ground:
  • Prior Art Mapping: Petitioner argued that Ophir discloses a secure network data storage "mediator" that acts as a secure storage system. This mediator provides "secure virtual storage" to clients, connects to multiple physical storage devices, and encrypts data before storage. However, Ophir describes encryption generally. Takeda was argued to supply the missing limitations by teaching an improved ciphering technique where plaintext is divided into blocks, each block is encrypted with a different "ciphering attribute" (e.g., a key), and the resulting ciphertext blocks are stored with their corresponding attributes or a pointer to them. The combination allegedly meets all limitations of the challenged claims, including generating shares (Takeda's ciphertext blocks), using a securing operation (Takeda's ciphering), and including data indicative of a session key with each share (Takeda's stored attributes).
  • Motivation to Combine: A POSITA would combine Takeda's improved ciphering method with Ophir's secure storage system for at least three reasons: (1) to increase the reliability and security of Ophir’s system, as Takeda explicitly teaches that if one part of the data is compromised, the rest is not easily deciphered; (2) as a simple substitution of Takeda’s known, more secure encryption method for Ophir’s generalized encryption to achieve predictable results; and (3) as the use of a known technique to improve a similar device in the same way.
  • Expectation of Success: A POSITA would have had a reasonable expectation of success because Takeda provides sufficient detail to implement its ciphering techniques into Ophir’s encryption/decryption unit. The components were compatible, as both systems were directed to encrypting and storing data securely.

Ground 2: Obviousness over Dickinson and Takeda - Claims 1-10 are obvious over Dickinson in view of Takeda.

• Prior Art Relied Upon: Dickinson (WO 2001/022322) and Takeda (Application # 2001/0053221).

• Core Argument for this Ground:

  • Prior Art Mapping: Petitioner asserted that Dickinson, which was the primary reference applied during prosecution, discloses nearly all limitations of the claims. Dickinson teaches a "data splitting process" that creates independently undecipherable portions (shares) of sensitive data (a dataset) using an XOR operation with random numbers (session keys) and distributes them among multiple data storage facilities. The Examiner found Dickinson did not explicitly teach including "data indicative of the at least one session key" with each share. Petitioner argued that Takeda, a reference not before the Examiner, remedies this by explicitly teaching the storage of ciphering attributes (which include key length or the key itself) or a pointer to the key with each block of encrypted data.
  • Motivation to Combine: Dickinson's system must retrieve both an encrypted data portion and its corresponding key to decrypt it but does not specify how the system associates the two. A POSITA would have been motivated to incorporate Takeda's technique of storing a pointer or metadata with the encrypted data. This provides a known and beneficial method for creating the necessary association, allowing for efficient retrieval and decryption while enabling separate management of keys and data.
  • Expectation of Success: A POSITA would have reasonably expected success in applying Takeda's pointer/metadata techniques to Dickinson's data-splitting process. Storing metadata with encrypted data was a common technique, and pointers were a known software tool for identifying data locations, making the modification straightforward.

• Additional Grounds: Petitioner asserted that claim 7 is obvious over Ophir, Takeda, and Birrell (Patent 7,529,834), where Birrell teaches data redundancy to allow reconstitution from fewer than all shares. Petitioner also asserted that claim 9 is obvious over Ophir, Takeda, and Dickinson, where Dickinson provides a specific example of data "splitting" using an XOR operation.

4. Key Claim Construction Positions

• The petition highlighted a dispute over the term "data indicative of the at least one session key" from the co-pending district court litigation.
• Petitioner (Google) proposed a construction of "metadata about [the] key[]; not an encryption key."
• Patent Owner (SFI) proposed a construction covering "information that points out, points to, makes known, shows, directs to, hints at, or suggests..." the key.
• Petitioner argued its obviousness contentions render the claims unpatentable under either construction. For example, Takeda's disclosure of storing key length with an encrypted block meets Google's construction, while Takeda's disclosure of storing the key itself or a pointer to the key meets SFI's construction.

5. Arguments Regarding Discretionary Denial

• Petitioner argued against discretionary denial under Fintiv (§314(a)), asserting compelling evidence of unpatentability. It noted that the district court trial was scheduled for July 2024, that the petition was filed diligently (three months after the patent owner identified asserted claims), and that there is limited issue overlap, as the petition challenges ten claims while only six were asserted in litigation.
• Petitioner also argued against denial under §325(d), stating that Grounds 1 and 2 present new art (Ophir, Takeda, Birrell) not considered during prosecution. For Grounds 3 and 4, Petitioner contended that Takeda supplies the very limitation that the Examiner relied upon for allowance over Dickinson, and the Examiner did not have the benefit of Takeda's teachings.

6. Relief Requested

• Petitioner requests institution of inter partes review and cancellation of claims 1-10 of the ’140 patent as unpatentable.