Technophile NewsTechnophile News
  • Home
  • News
  • PC
  • Phones
  • Android
  • Gadgets
  • Games
  • Guides
  • Accessories
  • Reviews
  • Spotlight
  • More
    • Artificial Intelligence
    • Web Stories
    • Press Release
What's On

Gemini Rolls Out to Wear OS Smartwatches With Support for Natural Language Commands

10 July 2025

OpenAI Reportedly Working On an AI-Powered Web Browser, to Compete With Google Chrome

10 July 2025

Apple Reportedly Working on Sealed-Device Update System for MacBook Units at Apple Stores

10 July 2025

Tecno Pova 7 5G, Pova 7 Pro 5G With Dimensity 7300 Ultimate SoC Sale in India Starts Today: Price, Offers

10 July 2025

This Sony Soundbar System Sounds Like a Theater and It’s Cheaper Than Ever for Prime Day

10 July 2025
Facebook X (Twitter) Instagram
  • Privacy
  • Terms
  • Advertise
  • Contact Us
Thursday, July 10
Facebook X (Twitter) Instagram YouTube
Technophile NewsTechnophile News
Demo
  • Home
  • News
  • PC
  • Phones
  • Android
  • Gadgets
  • Games
  • Guides
  • Accessories
  • Reviews
  • Spotlight
  • More
    • Artificial Intelligence
    • Web Stories
    • Press Release
Technophile NewsTechnophile News
Home » A New Kind of AI Model Lets Data Owners Take Control
News

A New Kind of AI Model Lets Data Owners Take Control

By News Room9 July 20253 Mins Read
Facebook Twitter Pinterest LinkedIn Telegram Tumblr Reddit WhatsApp Email
Share
Facebook Twitter LinkedIn Pinterest Email

A new kind of large language model, developed by researchers at the Allen Institute for AI (Ai2), makes it possible to control how training data is used even after a model has been built.

The new model, called FlexOlmo, could challenge the current industry paradigm of big artificial intelligence companies slurping up data from the web, books, and other sources—often with little regard for ownership—and then owning the resulting models entirely. Once data is baked into an AI model today, extracting it from that model is a bit like trying to recover the eggs from a finished cake.

“Conventionally, your data is either in or out,” says Ali Farhadi, CEO of Ai2, based in Seattle, Washington. “Once I train on that data, you lose control. And you have no way out, unless you force me to go through another multi-million-dollar round of training.”

Ai2’s avant-garde approach divides up training so that data owners can exert control. Those who want to contribute data to a FlexOlmo model can do so by first copying a publicly shared model known as the “anchor.” They then train a second model using their own data, combine the result with the anchor model, and contribute the result back to whoever is building the third and final model.

Contributing in this way means that the data itself never has to be handed over. And because of how the data owner’s model is merged with the final one, it is possible to extract the data later on. A magazine publisher might, for instance, contribute text from its archive of articles to a model but later remove the sub-model trained on that data if there is a legal dispute or if the company objects to how a model is being used.

“The training is completely asynchronous,” says Sewon Min, a research scientist at Ai2 who led the technical work. “Data owners do not have to coordinate, and the training can be done completely independently.”

The FlexOlmo model architecture is what’s known as a “mixture of experts,” a popular design that is normally used to simultaneously combine several sub-models into a bigger, more capable one. A key innovation from Ai2 is a way of merging sub-models that were trained independently. This is achieved using a new scheme for representing the values in a model so that its abilities can be merged with others when the final combined model is run.

To test the approach, the FlexOlmo researchers created a dataset they call Flexmix from proprietary sources including books and websites. They used the FlexOlmo design to build a model with 37 billion parameters, about a tenth of the size of the largest open source model from Meta. They then compared their model to several others. They found that it outperformed any individual model on all tasks and also scored 10 percent better at common benchmarks than two other approaches for merging independently trained models.

The result is a way to have your cake—and get your eggs back, too. “You could just opt out of the system without any major damage and inference time,” Farhadi says. “It’s a whole new way of thinking about how to train these models.”

Share. Facebook Twitter Pinterest LinkedIn Tumblr Email

Related News

This Sony Soundbar System Sounds Like a Theater and It’s Cheaper Than Ever for Prime Day

10 July 2025

McDonald’s AI Hiring Bot Exposed Millions of Applicants’ Data to Hackers Who Tried the Password ‘123456’

10 July 2025

Elon Musk Unveils Grok 4 Amid Controversy Over Chatbot’s Antisemitic Posts

10 July 2025

Musk makes grand promises during Grok 4 live demo, ignores Nazi scandal

10 July 2025

Our Favorite TV for Most People Is at Its Lowest Price Ever on Prime Day

10 July 2025

This Camera-Equipped Automatic Litter Box Is a Prime Day Deal

10 July 2025
Top Articles

Huawei Nova 14 Ultra – Price in India, Specifications (21st May 2025)

20 May 2025125 Views

iQOO Neo 10 Pro+ Confirmed to Debut This Month, Pre-Reservations Begin

8 May 202589 Views

Redmi K80 Ultra Design, Colours, and Key Features Revealed; to Get MediaTek Dimensity 9400+ SoC

18 June 202580 Views
Stay In Touch
  • Facebook
  • YouTube
  • TikTok
  • WhatsApp
  • Twitter
  • Instagram
Don't Miss

Apple Vision Pro Successor to Reportedly Launch in 2025 With M4 Chip, Better Components for AI

10 July 2025

Apple is rumoured to be developing a successor to the Vision Pro and it could…

Samsung Galaxy S25 FE Design Appears on WPC Database; May Support Qi2 Wireless Charging

10 July 2025

McDonald’s AI Hiring Bot Exposed Millions of Applicants’ Data to Hackers Who Tried the Password ‘123456’

10 July 2025

Apple Support App Reportedly Getting a ChatGPT-Style AI Chatbot to Assist Users

10 July 2025
Technophile News
Facebook X (Twitter) Instagram Pinterest YouTube Dribbble
  • Privacy Policy
  • Terms of use
  • Advertise
  • Contact Us
© 2025 Technophile News. All Rights Reserved.

Type above and press Enter to search. Press Esc to cancel.