The Forge

Powering liquid markets for digital human expertise

While ChatGPT can think and write, the next evolution of AI will actually DO — controlling your computer, executing tasks, automating entire workflows. This requires training data of humans actually using computers, not just text conversations.

The Forge is where businesses & individuals create "Factories" to collect high-quality human demonstrations for training Computer Use Agents. Our production-ready desktop app captures every click, keystroke, and decision as you execute workflows — instantly converting your expertise into AI training datasets powering the first liquid infrastructure for computer-use training data.


Core Concept

Factories = Training Data Collection Environments

A Factory is a funded workspace that incentivizes contributors to record their screen while executing specific computer tasks to create valuable tokenized training datasets.


How It Works

1. Create a Factory

Design the data collection space for your training data. Name your Factory and define the workflow skills you want to capture. Choose a reward token ($CLONES/$USDC/$ETH).

2. Auto-Structure Training Tasks

Provide a skill prompt. The Forge transforms it into structured actionable tasks for contributors to demonstrate. Each task is optimized to be replayable, verifiable, and relevant for high-quality training data.

3. Fund Your Factory Treasury

Add your reward token to your Factory's treasury to attract contributors. Define the specific amount of reward per task — the larger the reward pool, the stronger the incentive. Dynamic quality analysis always ensures that the best demonstrations receive the most value.

4. Collect Demonstrations & Build Your Dataset

Contributors earn rewards by recording their screens while executing tasks from your Factory. Each recording is automatically processed, labeled, and quality-scored. All validated demonstrations are compiled into your Factory Dataset, creating valuable training data for computer-use agents.

5. Tokenize Your Training Dataset

Once your dataset reaches your desired amount of demonstrations:

Dataset Token Launch (Tokenization)

  • Launch via bonding curve (Launch fee: $CLONES + $ETH)

  • Set IP access threshold — amount of token to burn for IP access

  • Liquidity is migrated to DEX & burnt when token graduates at 69k MC

Revenue Streams (Monetization)

  • Trading Revenue: Earn 0.25% from all trading volume forever

  • Threshold Control: You set burn-to-download supply amount for IP access

  • Meta-Dataset Inclusion: Private sale allocations in elite datasets if your data is used


The Multiplier Effect:

  1. Your expertise gets captured once

  2. Your dataset becomes a tradeable token

  3. Token appreciates as enterprises adopt your workflows

  4. You earn 0.25% volume fees on every trade

  5. Meta-dataset inclusion creates additional token allocations

  6. Network effects amplify value across the ecosystem

"Your Factory evolves from a data collection workspace into a tokenized revenue-generating training data business"

Last updated