Google’s ‘Project Ellmann’ builds on Gemini AI to create a story of your life


What you need to know

  • Google details “Project Ellmann,” an AI model designed to be “Your Life Story Teller.”
  • It can collect information from your stored photos to detail who you haven’t seen in a while, pets, children, eating habits, favorite websites, and more.
  • The project, during its infancy, is also packed with “Ellmann Chat,” a chatbot where users can ask any question about their lives.

Google details a new AI project that could turn your saved memories into a complete story packed with background information.

The idea is to give users a “bird’s eye” view into their lives through “Project Ellmann,” CNBC reports. From materials the publication viewed, the new AI model would piggyback off LLMs like Gemini to “answer previously impossible questions.” It would collect information from a user’s search results, notice patterns in their photos, and contain a chatbot to act upon such interactions.

The project in question is also said to have the ability to grab information from biographies, past photographed moments, and new moments that follow to tell a story. A Google product manager stated during the presentation that Project Ellmann could discern the date of birth of a child, while also detailing their parents, and if they are an only child.

“Ellmann Chat” is the other side of this software, with Google stating, “Imagine opening ChatGPT but it already knows everything about your life. What would you ask it?”

With all of your memories at its disposal, the AI can answer questions such as “When was the last time I saw my sibling?” or “Do I have a pet?” Regarding the last question, Google states that the AI can provide specific details such as the name of your pet and who they were photographed with the most.

Of course, this sounds quite invasive, but the company explains that privacy is a deep concern. “This was an early internal exploration and, as always, should we decide to roll out new features, we would take the time needed to ensure they were helpful to people, and designed to protect users’ privacy and safety as our top priority,” a statement from a Google Spokesperson reads.

Project Ellmann can also understand a user’s eating habits (favorite foods, etc.) while gaining insight into their favorite websites and apps. With interest in creating an AI that can be “Your Life Story Teller,” it’s not clear where Google would roll Project Ellmann into. However, a product manager for Google Photos presented such information, which could give us an inkling as to where we may see this in the future.

(Image credit: Google)

Speaking of Google Photos, the app is no stranger to AI features as the company rolled out the “Memories” view earlier this year. Generative AI was at its core as the feature could curate and organize your memories alongside title suggestions for unique collections. There’s also the AI video editor that arrived back in October as a way to make quick and easy highlight reels to share.

Meanwhile, Google has been increasingly ramping up its efforts with AI models as it recently debuted Gemini — a large language model Project Ellmann could utilize. Within Gemini are three variants: Nano, Pro, and Ultra, all designed for various uses. While Pro excels at a wide range of tasks and Ultra for complex tasks, Nano’s expertise is on-device tasks.

If Google Photos is the aim of such software, would be fair to speculate Project Ellmann would tap Nano — if it survives the rigorous internal testing. Regardless, Google’s December 2023 feature drop brought Gemini Nano to the Pixel 8 series to be used for Recorder summaries and “Smart Reply in Gboard.”

Via: Androidcentral

Leave a Comment