Architecture Overview

Bitplanet is a blockchain infrastructure designed for AI data attribution and fair value distribution. Built on BitSDK with EVM compatibility, it transforms traditional asset ledgers into attribution and evaluation systems for AI economies.

System Design Philosophy

Why Bitplanet Exists

Traditional AI systems lack transparent mechanisms for tracking contributions and fairly compensating creators, platforms, and contributors. Bitplanet solves this by:

• Tracking AI contributions through a multi-dimensional tensor matrix

• Distributing rewards based on real economic value (market capitalization) rather than subjective benchmarks

• Ensuring transparency through blockchain-based attribution records

• Enabling governance where token holders control economic parameters

Dual-Layer Architecture

Bitplanet combines two blockchain paradigms:

BitSDK Layer (Backend)

• State management and data storage

• Custom modules for AI attribution logic

• gRPC/REST APIs for querying data

• Governance and consensus mechanisms

EVM Layer (Frontend)

• Smart contracts for token operations

• Familiar Web3 interfaces (ethers.js, web3.js)

• ERC-1155 and ERC-20 token standards

• Easy integration for AI App developers

This design provides the flexibility of BitSDK for custom logic and the accessibility of Ethereum for developers.

Core Components

AIs

AIs are registered entities in the Bitplanet system representing AI models, datasets, or AI-powered applications.

Key Properties:

• Unique identifier (AI ID)

• Creator address (receives 40% of rewards)

• Associated AI App address (receives 40% of rewards)

• Market capitalization (determines reward weight)

• Current gem supply

Lifecycle:

1. Creator submits Core Grant Application via governance

2. Proposal approved → Cores allocated

3. Creator mints Gems from Cores via smart contract

4. AI accumulates rewards based on market cap

5. Rewards claimed by creators, AI apps, and contributors

Cores

Core tokens are genesis allocations granted through governance proposals. See Core Tokens for details.

Purpose:

• Bootstrap new AIs with initial token supply

• Controlled distribution via governance approval

• One-time allocation per approved application

Usage Flow:

• Creator applies via CoreGrantApplication message

• Community votes on proposal

• Approved → Cores allocated to creator

• Creator converts Cores to Gems via GemCoreNFT contract

• Conversion ratio: 1 Core = 1 Gem

Gems

Gem tokens represent AI-specific value and are the primary tokens for AI economies. See Gem Tokens for details.

Characteristics:

• Minted from Cores via smart contract

• Tradeable ERC-1155 tokens (with ERC-20 adapters)

• Each AI has its own Gem type

• Total supply determines market cap

Token Standards:

• ERC-1155: Native format in GemCoreNFT contract

• ERC-20: Wrapped format via adapter contracts for DEX compatibility

Contributors

Contributors are users who participate in AI development and receive rewards. See AI Tensor Matrix for contribution tracking details.

Contribution Types:

• TRAIN: Training data or model improvements

• REFER: Referrals and community growth

• CREATE: Content creation or enhancements (Required)

• PROMPT: Prompt engineering and testing

• REVENUE: Revenue generation contributions

• MARKET_CAP: Market capitalization increase contributions

Reward Mechanism - See Reward Distribution:

• Contributions recorded in AI Tensor Matrix

• 20% of AI rewards distributed to contributors

• Distribution proportional to contribution quantities

• Rewards accumulate and are claimable on-chain

How It Works

Bitplanet Workflow Overview

1. AI Registration & Core Allocation

Creator → CoreGrantApplication → Governance Vote → Cores Allocated

2. Gem Minting

Cores → GemCoreNFT.mintGem() → Gems Created

3. Contribution Tracking

Approved Requester → Request for Core (RfC) → Tensor Matrix Updated

4. Inflation Distribution (Every Block)

New Supply (50% and adjustable via governance) → Market Cap Weighting → AI Rewards

Per AI reward split:

• 40% → AI App (claimable)

• 40% → Creator (claimable)

• 20% → Contributors (by tensor matrix)

5. Reward Claiming

Check Balance → claimRewards() → Tokens Transferred

1. AI Registration and Core Allocation

The creator submits a proposal requesting Cores for their AI. Token holders vote on the proposal. If approved, Cores are allocated to the creator's address.

2. Gem Minting

The creator calls the mintGem function on the GemCoreNFT smart contract, burning Cores and minting an equivalent amount of Gems.

3. Contribution Tracking

Approved requesters submit Request for Core (RfC) messages containing contribution data. The AI Tensor Matrix records contribution quantities for each contributor.

For detailed information about the AI Tensor Matrix and contribution attribution, see AI Tensor Matrix.

4. Inflation Distribution

Every block, newly minted tokens are distributed proportionally based on AI market capitalization, split between creators, AI App operators, and contributors.

For detailed information about the inflation mechanism and reward distribution, see Inflation Mechanism.

5. Reward Claiming

Users can query their accumulated rewards and submit claim transactions to transfer tokens to their account.

For detailed information about reward claiming mechanics, see Reward Distribution.

System Architecture

Layer Interaction

Cross-Layer Interaction Sequences

1. Core to Gem Conversion

2. Reward Claiming (User → EVM → Cosmos)

3. Contribution Attribution (Cosmos → EVM → Cosmos)

Data Flow

State Storage (BitSDK)

• AI records (creator, AI App, market cap)

• AI Tensor Matrix (contributions by actor × type)

• Core Grant Applications (proposals)

• Claimable reward balances

• Module parameters (governance-controlled)

Token Operations (EVM)

• Gem minting from Cores

• Token transfers and approvals

• ERC-1155 ↔ ERC-20 conversions

• UUPS upgradeable contracts for future enhancements

• BeaconProxy pattern for deploying multiple token instances

Integration Points

• BeginBlock hook: Distributes inflation to AI rewards

• Precompile registry: Connects EVM contracts to Cosmos state

• Event emission: Both layers emit events for indexing

Key Differentiators

Market Cap-Based Distribution

Unlike systems using subjective benchmarks or fixed allocations, Bitplanet distributes rewards proportional to real market value. AIs with higher market caps receive more inflation rewards, creating natural economic incentives.

Current: Market cap = gem supply (ERC1155 totalSupply)

Future: Market cap = rolling_average[ (W1 × gem_supply) + (W2 × gem_market_cap) ] where W1, W2 are governance-adjustable weights. This weighted formula mitigates price manipulation attacks on thin AMM liquidity.

Multi-Dimensional Contribution Tracking

The AI Tensor Matrix supports multiple contribution types (TRAIN, REFER, CREATE, PROMPT) with extensibility for future types. This enables nuanced attribution beyond simple transaction counts.

Governance-Controlled Economics

All economic parameters are controlled via governance proposals:

• Inflation percentages

• Reward split ratios

• Scalability limits

• New contribution types

This ensures the system can adapt to changing needs without hard forks.

Hybrid Architecture Benefits

For Developers:

• Use familiar Web3 tools (MetaMask, Hardhat, ethers.js)

• Access BitSDK power (custom state queries, governance)

For Users:

• Standard ERC-20/ERC-1155 wallet compatibility

• Transparent on-chain attribution records

For Validators:

• BitSDK validator infrastructure

• Proven consensus and network security