Aydın Tiryaki and Gemini AI (2025)

​Human relationships are more than just names and faces; they are the sum of shared moments. However, our minds occasionally struggle to retrieve the narrative behind a face, even when it feels familiar. As the second phase of the “Tip of My Tongue” project, “Where Do I Know Them From?” is a social navigation model designed to transform moments of social hesitation from embarrassment into a graceful technological experience.

​1. A Social Safety Net: Turning Familiarity into Information

​We have all experienced it: a smiling face approaching us, our brain signaling “I know them,” but the question “Who?” remains unanswered. This project positions AI as an “External Memory Unit” to fill this cognitive gap. The system provides the necessary “anchor information” silently, without disrupting the user’s social flow.

​2. The Experiential Filter and the Private Visual Vault

​The fundamental difference between “Where Do I Know Them From?” and ordinary facial recognition technologies is the “Experiential Filter.” This database is not built from cold data harvested from the internet; it is woven exclusively from the records of people the user has personally met, spoken with, or shared a space with.

• ​The Private Visual Vault: When an unrecognizable face is encountered, the system places the image in an encrypted “Visual Quarantine.” This image is anonymous; it is never transferred to cloud systems for training and remains inaccessible to anyone but the owner.
• ​Delayed Labeling: When the user eventually recalls the name (later that day or during a break), they define it to the system: “The person I saw today was my former colleague, Murat.”
• ​The Permanent Bridge: The system seals this information for the next encounter. During the second meeting, before the user even asks, “Where do I know them from?”, the system recognizes Murat and provides the context.

​3. Dual-Track Recognition: Public vs. Private

​To ensure cognitive energy efficiency, the system operates through two distinct channels:

• ​Public Figures (Celebrities): Authors, artists, or figures known to the public are instantly recognized from general AI knowledge banks.
• ​The Private Social Network: Only individuals within the user’s “Life Story” are stored in the private archive with high-level encryption.

​4. Sensory Preferences: Customized Accessibility for Everyone

​This project defines the need for support not as a “deficiency” but as a “User Experience Comfort.” Stripped of medical terminology, the system adapts to how the user prefers to “read” the world:

• ​Auditory Companion (Visual-to-Audio): For those who prefer to hear information, the system whispers the name and context gently via an earpiece (“Social Prompter”).
• ​Visual Support (Audio-to-Text): For those who prefer to read, the system displays the necessary notes on a smartwatch or phone screen like a “Live Social Caption.”

​5. Preserving Dignity: Security and Technical Armor

​Protecting human dignity and privacy is the highest priority. Operating through a Micro-App (Nano-App) architecture, the system uses AI solely as a “processor,” while the data ownership remains entirely with the user. In-app screenshot restrictions and sharing blocks make it physically impossible for this digital diary to be leaked.

​Conclusion

The “Where Do I Know Them From?” project is a “Cognitive Orthosis” that protects an individual’s social standing and bonds with loved ones, especially during the natural forgetfulness that may come with age. Turning a brief moment of silence into a warm embrace by whispering a name is the most noble service technology can offer to humanity.

Aydın Tiryaki and Gemini AI Ankara, December 27, 2025

A Note on Methods and Tools: All observations, ideas, and solution proposals in this study are the author’s own. AI was utilized as an information source for researching and compiling relevant topics strictly based on the author’s inquiries, requests, and directions; additionally, it provided writing assistance during the drafting process.

(The research-based compilation and English writing process of this text were supported by AI as a specialized assistant.)