Home
BlogCareers

Community

The Uncharted Data Challenge

·

Apr 02, 2026

·

2min read

The most valuable datasets in the world don't exist yet.

They're in languages that mainstream AI has never been trained on. They're in industries and centered on problems that have never had the research attention they deserve. They're sitting in the gaps and right now, nobody is building them.

Today, we’re launching the Uncharted Data Challenge, a two-week open data challenge with a $20,000 prize pool for the builders, researchers, and communities ready to fill those gaps.

Adapted to Serve the World

At Adaption, we believe the next frontier of AI isn't about scale. It's about reach. The teams doing the most consequential work are the ones building for industries and communities that mainstream AI has never been optimized for.

Data is where that work starts. And right now, the world is missing too much of it. The Uncharted Data Challenge is our invitation to the global community to help change that.

This challenge is powered by Adaptive Data, our first product. Adaptive Data is the data layer of the Adaption platform, designed to dynamically reshape data at scale as your product, users, and edge cases evolve. It is the first of three pillars we are building : Adaptive Data, Adaptive Intelligence and Adaptive Interfaces.

The Challenge

Participants are invited to create and open-source high-quality datasets across two weeks. Our research staff will be looking for three criteria:

  1. Underserved problem domains: datasets addressing areas with little existing research attention.
  2. Scarce open-source data: filling genuine gaps in publicly available data.
  3. Under-resourced languages: datasets that represent languages underrepresented in machine learning.

All submissions must be open-sourced to Kaggle or Hugging Face and must credit Adaptive Data by Adaption.

The Prize Pool

Total prize pool: $20,000 USD.

  • 1st Place - $12,000
  • 2nd Place - $5,000
  • 3rd Place - $3,000

How to Participate

Sign up to express your interest. Applications are reviewed on a rolling basis. Accepted participants will be contacted directly with next steps, credit allocation details, and access to Adaptive Data.

All accepted participants will receive Adaptive Data credits to use during the competition period. Credit allocation details will be confirmed at the time of acceptance.

This challenge is for anyone working at the intersection of data, language, and impact: researchers, engineers, linguists, domain experts, and community builders. If you've ever looked at the state of AI in your language or your industry and thought this isn't good enough, this challenge was made for you.

Challenge Opens: Friday, April 10, 2026

Challenge Closes: Friday, May 1, 2026

Winners Announced: Tuesday, May 5, 2026

Author

Sara Hooker, Co-founder and Sudip Roy, Co-founder

Date

Apr 02, 2026