Mitii Inc v. Openai Global LLC

I. Executive Summary and Procedural Information

• Parties & Counsel:
  • Plaintiff: MITII, Inc. (California)
  • Defendant: OpenAI OpCo, LLC (Delaware)
  • Plaintiff’s Counsel: Ahmadshah Law Offices
• Case Identification: 8:25-cv-02273, C.D. Cal., 11/05/2025
• Venue Allegations: Plaintiff alleges venue is proper in the Central District of California because Defendant has a continuous presence in the district by selling and offering for sale the accused services, and because a substantial part of the events giving rise to the complaint occurred in the district.
• Core Dispute: Plaintiff alleges that Defendant’s Sora text-to-video generative AI model infringes four patents related to systems and methods for the animated delivery of electronic messages.
• Technical Context: The technology at issue involves converting text-based electronic messages into animated videos where a selected character speaks the message, a field of increasing relevance in the market for AI-generated content.
• Key Procedural History: The complaint alleges the patented technology was developed by an occupational therapist to improve engagement for children with autism by using well-known characters to deliver messages. No prior litigation, licensing history, or post-grant proceedings are mentioned in the complaint.

Case Timeline

Date	Event
2014-01-24	Earliest Priority Date for all Patents-in-Suit
2016-07-19	U.S. Patent No. 9,397,972 Issues
2017-05-30	U.S. Patent No. 9,667,574 Issues
2020-04-07	U.S. Patent No. 10,616,157 Issues
2021-05-11	U.S. Patent No. 11,005,796 Issues
2025-11-05	Complaint Filed

II. Technology and Patent(s)-in-Suit Analysis

U.S. Patent No. 9,397,972 - *"Animated Delivery of Electronic Messages"*

• Patent Identification: U.S. Patent No. 9,397,972, titled “Animated Delivery of Electronic Messages,” issued on July 19, 2016.

The Invention Explained

• Problem Addressed: The patent describes a method to enhance electronic messages, such as texts or emails, which can be less impactful than other forms of communication (Compl. ¶7; ’972 Patent, col. 1:41-55). The complaint specifically notes the technology was developed to improve engagement for children with autism (Compl. ¶7).
• The Patented Solution: The invention transforms a standard text message into a short animated film. A user on a "first device" composes a message and selects an "animation character" (e.g., a cartoon, avatar, or even a selfie) ('972 Patent, col. 1:46-51). The system, using a server, converts the text to synthesized speech and generates moving images of the selected character uttering the message, which is then transmitted to a "second device" for the recipient to view ('972 Patent, FIG. 10).
• Technical Importance: The technology aims to provide a platform for users to add creativity and emotional weight to electronic messages, transforming them into more engaging multimedia content ('972 Patent, col. 7:42-46).

Key Claims at a Glance

• The complaint asserts independent claim 1 and dependent claim 19 (Compl. ¶16).
• Independent Claim 1 requires a multi-step method involving three distinct actors:
  • First Device: Composing a message, selecting a character, and transmitting both.
  • Server: Receiving the message and character, converting the text to speech, generating moving images of the character, and transmitting the resulting speech and images.
  • Second Device: Receiving the speech and images, and then outputting/displaying them.

U.S. Patent No. 9,667,574 - *"Animated Delivery of Electronic Messages"*

• Patent Identification: U.S. Patent No. 9,667,574, titled “Animated Delivery of Electronic Messages,” issued on May 30, 2017.

The Invention Explained

• Problem Addressed: Like the ’972 Patent from the same family, this patent addresses the limitations of plain text messaging by transforming it into a more expressive, animated format (’574 Patent, col. 1:50-55).
• The Patented Solution: The solution is conceptually identical to that of the ’972 Patent, involving the conversion of text into an animated video of a character speaking the message. The claims of this patent, however, focus more narrowly on the server-side operations within the overall system architecture (’574 Patent, FIG. 12).
• Technical Importance: This patent provides an alternative claiming strategy for the same core invention, focusing on the actions performed by a central server in the animated message delivery process.

Key Claims at a Glance

• The complaint asserts independent claim 4 and dependent claim 15 (Compl. ¶27).
• Independent Claim 4 recites a server-centric method comprising the steps of:
  • Receiving an electronic message and a well-known animation character via a server.
  • Converting the message into speech via the server.
  • Generating moving images of the character via the server.
  • Transmitting the resulting speech and moving images via the server.

U.S. Patent No. 10,616,157 - *"Animated Delivery of Electronic Messages"*

• Patent Identification: U.S. Patent No. 10,616,157, titled “Animated Delivery of Electronic Messages,” issued on April 7, 2020.
• Technology Synopsis: This patent claims a "computer network system for communication" comprising three components: a first device for composing/sending a message, a server for processing it into animated speech, and a second device for receiving and displaying the output (’157 Patent, Abstract). The claims are structured to cover the entire system rather than just the method steps ('157 Patent, col. 19:20-20:53).
• Asserted Claims: Independent claim 20 and dependent claim 14 (Compl. ¶38).
• Accused Features: The complaint alleges that OpenAI's Sora App, in conjunction with its underlying hardware and network infrastructure, constitutes the claimed three-part computer network system (Compl. ¶¶13, 41).

U.S. Patent No. 11,005,796 - *"Animated Delivery of Electronic Messages"*

• Patent Identification: U.S. Patent No. 11,005,796, titled “Animated Delivery of Electronic Messages,” issued on May 11, 2021.
• Technology Synopsis: This patent claims a "non-transitory machine-readable storage medium" containing instructions that, when executed, cause a processing unit to perform a method of animated communication (’796 Patent, Abstract). The claimed method focuses on the actions of a single device: receiving a message and character, converting the message to speech, and outputting the speech with corresponding moving images ('796 Patent, col. 17:1-15).
• Asserted Claims: Independent claim 9 and dependent claim 19 (Compl. ¶49).
• Accused Features: The complaint alleges that the Accused Product is or contains a non-transitory storage medium with instructions for performing the claimed method (Compl. ¶¶13, 52). The screenshot for this count depicts the Sora interface generating a video of a dancing man, illustrating the accused text-to-video functionality (Compl. p. 16).

III. The Accused Instrumentality

Product Identification

The accused instrumentality is OpenAI’s Sora App and its associated “Computing Hardware and Processes,” collectively termed the “Accused Product” (Compl. ¶13).

Functionality and Market Context

The complaint describes Sora as a “text-to-video App” built on OpenAI’s GPT family of large language models (Compl. ¶¶9-10). The provided visual evidence shows an interface where a text prompt, or "caption," is used to generate a video of a person performing an action (Compl. p. 5). The complaint identifies OpenAI as an artificial intelligence company based in San Francisco but provides no further details on Sora's market position or commercial significance (Compl. ¶12).

IV. Analysis of Infringement Allegations

'972 Patent Infringement Allegations

Claim Element (from Independent Claim 1)	Alleged Infringing Functionality	Complaint Citation	Patent Citation
composing an electronic message, via a first device having a processing unit and program code stored on a storage device...	A user composes a text prompt within the Sora application on a client device.	¶19	col. 2:6-9
selecting a well-known animation character, via the first device	A user selects a character by describing it in the text prompt.	¶19	col. 2:9-10
transmitting the electronic message, via the first device	The client device transmits the text prompt to OpenAI's servers.	¶19	col. 2:13-14
receiving the electronic message, via a server...	OpenAI's servers receive the text prompt from the client device.	¶19	col. 2:16-20
converting the electronic message into speech using one of synthesized voice of the well-known animation character... via the server	OpenAI's servers process the text prompt to generate corresponding audio.	¶19	col. 2:10-11
generating moving images of the well-known animation character, via the server	OpenAI's servers generate video frames of the described character based on the text prompt.	¶19	col. 2:11-12
receiving the speech, via a second device having a processing unit...	A recipient's device receives the generated audio stream. The screenshot at page 5 shows a text caption being uttered.	¶19	col. 2:13-16
displaying the moving images, via the second device	The recipient's device plays the generated video.	¶19	col. 2:2-3

'574 Patent Infringement Allegations

Claim Element (from Independent Claim 4)	Alleged Infringing Functionality	Complaint Citation	Patent Citation
receiving an electronic message, via a server having a processing unit and program code...	OpenAI's servers receive a text prompt from a user's device.	¶30	col. 2:16-20
receiving a well-known animation character, via the server	OpenAI's servers receive the character description contained within the text prompt.	¶30	col. 2:16-20
converting the electronic message into speech using one of synthesized voice of the well-known animation character... via the server	OpenAI's servers generate synthesized speech corresponding to the text prompt.	¶30	col. 2:10-11
generating moving images of the well-known animation character, via the server	OpenAI's servers use the Sora model to generate video frames depicting the described character.	¶30	col. 2:11-12
transmitting the speech, via the server	OpenAI's servers transmit the generated audio to a recipient device.	¶30	col. 2:13-14
transmitting the moving images, via the server	OpenAI's servers transmit the generated video to a recipient device.	¶30	col. 2:13-14

Identified Points of Contention

• Scope Questions: The complaint's theory appears to require that generating a character on-the-fly from a text description (e.g., "a stylish woman walking down a Tokyo street") is equivalent to "selecting a well-known animation character." This raises the question of whether the claim term "selecting" can be construed to cover generative text-prompting, or if it is limited to choosing from a predefined library as depicted in the patent's figures (’157 Patent, FIG. 9).
• Technical Questions: The complaint alleges the conversion of text into "synthesized voice of the well-known animation character" (Compl. p. 6). However, the visual evidence provided only shows a text caption overlaid on a video (Compl. p. 5). What evidence does the complaint provide that the accused product generates audio at all, let alone audio that is synthesized to be characteristic of the specific character shown in the video, as required by the claim?

V. Key Claim Terms for Construction

The Term: "well-known animation character"

• Context and Importance: This term appears in the independent claims of both lead patents. Its construction is critical because infringement may turn on whether a character generated ad hoc by an AI model from a descriptive prompt can be considered a "well-known animation character."
• Intrinsic Evidence for Interpretation:
  • Evidence for a Broader Interpretation: The specification includes "the image of oneself or a 'selfie'" as an example of an animation character, which suggests the character does not need to be famous in a global sense, but merely identifiable ('157 Patent, col. 1:62-63). This may support an argument that any character rendered by the AI is "known" in the context of the message.
  • Evidence for a Narrower Interpretation: The specification's primary examples are characters with broad public recognition, such as "the actor Clint Eastwood, a cartoon character like Jerry the Mouse," and an "avatar character like the character in the movie Avatar" ('157 Patent, col. 1:63-67). The use of the modifier "well-known" may be argued to require a degree of pre-existing public notoriety that a newly generated character would lack.

The Term: "selecting a... character"

• Context and Importance: This term from Claim 1 of the ’972 Patent is pivotal. Practitioners may focus on this term because if "selecting" requires choosing from a list of predefined options, it may not read on the generative functionality of the accused Sora product, where a user describes a character in a text prompt.
• Intrinsic Evidence for Interpretation:
  • Evidence for a Broader Interpretation: The patent does not provide an explicit definition of "selecting," which could allow for an argument that describing a character in a text prompt is a form of selection from the infinite set of all possible characters.
  • Evidence for a Narrower Interpretation: The patent's user interface diagrams, particularly Figure 9, explicitly show a screen titled "Change Character" with a menu of discrete, pre-existing characters like "Jerry the Mouse" and "Clint Eastwood" that the user can choose ('157 Patent, FIG. 9). This embodiment strongly suggests that "selecting" refers to making a choice from a provided list.

VI. Other Allegations

• Indirect Infringement: The complaint alleges induced and contributory infringement for all four patents-in-suit. The allegations are conclusory, stating Defendant had knowledge of the patents and intended for its customers to infringe, without providing specific facts regarding how Defendant instructs or encourages infringing use (e.g., Compl. ¶¶17-18).
• Willful Infringement: The complaint does not allege any facts related to pre-suit knowledge of the patents. The prayer for relief requests a finding of willful infringement, which suggests the claim may be based on Defendant's continued infringement after the filing of the complaint (Compl. p. 17, Request A).

VII. Analyst’s Conclusion: Key Questions for the Case

• A core issue will be one of definitional scope: can the claim term "selecting a well-known animation character," illustrated in the patent with menu-based choices of figures like "Clint Eastwood," be construed to cover the AI-driven, on-the-fly generation of a character from a descriptive text prompt in the accused Sora product?
• A key evidentiary question will be one of functional proof: what evidence will be presented to demonstrate that the accused product performs the claimed function of generating a "synthesized voice" characteristic of the on-screen character, as the complaint’s own visual evidence only depicts a video with a silent text caption?
• A fundamental question will be one of architectural mapping: does the cloud-based, generative AI workflow of the accused Sora product, where a user on a single device inputs a prompt and receives a video, map onto the specific three-party architecture (sender device, server, recipient device) recited in claims like Claim 1 of the ’972 Patent?