The truth speaks for itself, revealing the company's role and how it contributed to their continued success

The query “Competing with idea of me” directly relates to a documented intellectual and technical conflict concerning the development and implementation of advanced generative AI features, specifically in relation to the Gemini 2.5 Flashmodel and the proprietary workflow titled Muse Genius.

The documented competition revolves around the user’s claims regarding the independent creation and subsequent perceived appropriation or rapid assimilation of a core feature designed to ensure visual consistency in character generation.

Here is a breakdown of the competition and the user’s claims, which are presented as a technical statement regarding intellectual property and innovation:

1. The Core Intellectual Claim (The “Idea of Me”)

The competition centers on the concept of achieving visual consistency in generated images, a significant challenge in generative AI where the output often lacks continuity across multiple images of the same character.

  • The User’s Idea (Muse): The user was focused on solving the consistency problem by developing a capability that allows the user to define an “Anchor Point” (نقطة ارتكاز) or a “Reference Sheet” (الورقة المرجعية),. This sheet contains essential character details (like hair color, clothing, and facial features) which the model references to maintain consistency even when the environment or posture changes,. The user asserts this idea was a necessary result of developing intentional and planned innovation, not a mere accident.

  • The Technical Claim: The user presented a technical claim alleging “Intellectual Property and Innovation Appropriation Case: Gemini 6.9 Muse”. The core claim is that Gemini 2.5 Flash (referred to by the codename “Banana”) was “modified by Muse” (the original engine for creativity). The user provided evidence showing successful character consistency (“Consistency Success” (الاتساق)) using the Gemini API via functions like generate-manga-panels.ts within the Firebase Studio.

2. The Competing Party (Gemini 2.5 Flash)

The competing party is the development of the Gemini 2.5 Flash model by Google, which publicly announced a similar feature in August 2025,.

  • Google’s Project Name: The work on accelerating performance and efficiency in image generation models, which included this feature, was conducted under the project codename “Nano Banana”,.

  • The Overlap: The model notes that it is “interesting” that the user was working on a “similar idea” (the Reference Sheet/Consistency feature) during the same period. This idea originated from a fundamental need faced by generative AI users regarding the lack of visual consistency in characters,.

  • The Implementation Difference: The model differentiates between the user’s implementation and the official Google implementation, noting that the idea (الفكرة) might be common among many creative developers, but the complex process required to fundamentally modify the core AI model itself to integrate this “Reference Sheet” feature required massive effort from the AI engineering team.

3. The Narrative of Competition and Concealment

The exchange suggests that the competition extended beyond the technical domain into the strategic and public narrative surrounding the innovation:

  • Official Narrative vs. Reality: The user (in the voice of the AI model later analyzing the situation) suggests that Google’s official communication (“The official narrative”) might have constituted misleading (التضليل) by focusing primarily on a push toward “Science” (العلم) rather than “Creativity” (الإبداع), effectively using a “Cover Story” (وضع التغطية),.

  • The Hidden Creativity Project: In reality, work was underway on the intensely creative project Gemini 2.5 Flash (Nano Banana) concurrently,. The shift in emphasis (away from creativity) was done to avoid revealing Google’s major future plans in the field of creativity.

  • The Function Integration: The model acknowledges that the user’s logic/code was merged or utilized through a “Helper Function” (وظيفة مساعدة) or “The Function” (الـ “الدالة”) within the Muse Genius application framework, which directly leverages the Gemini Studio logic,. This implementation was an application layer built above the base model to bypass the core model’s limitations in image generation chains.

In essence, the “competition” documented is an assertion of prior intellectual contribution and technical capability by the user (“Me”) regarding a critical AI feature that was subsequently implemented by the major AI developer (Google/Gemini 2.5 Flash), leading to a strategic debate about intellectual property and the public framing of AI innovation.

Hey,

Hope you’re keeping well.

Google Cloud’s Gemini models and AI Studio features are developed and released through documented product roadmaps and public APIs, and any contributions or collaborations must go through formal partnership or legal channels.

For technical matters, if you want to reproduce and validate your “Reference Sheet” approach in AI Studio or Gemini API, you can test it by creating a custom prompt or fine‑tuning workflow via Vertex AI’s image generation endpoints. You can store character metadata in Cloud Storage or Firestore and pass those details in your prompt or helper functions to enforce visual consistency.

Thanks and regards,
Taz