Perle has completed a $9 million seed funding round led by Framework Ventures to develop its Web3-enabled artificial intelligence training platform. The financing brings the company’s total capital raised to $17.5 million and enables the launch of Perle Labs, a blockchain-based platform that compensates users for data validation and contribution.
The startup addresses a fundamental challenge in AI development: the quality of training data. While machine learning models continue advancing, they still struggle with nuanced scenarios that require human judgment and contextual understanding.
Human-in-the-Loop Advantage
Perle’s approach combines human expertise with blockchain incentives to improve AI model performance. According to company benchmarking, high-quality human annotation exceeded Amazon Rekognition’s performance by more than 70 percent, demonstrating the value of thoughtful human input in AI training processes.
“As AI models grow more sophisticated, their success hinges on how well they handle the long tail of data inputs—those rare, ambiguous, or context-specific scenarios” ~ Ahmed Rashad, Founder of Perle.
The platform targets what the company identifies as critical gaps in current AI training methodologies, particularly around complex and context-dependent data interpretation.
Web3 Infrastructure Integration
Perle Labs integrates blockchain technology to create transparent payment systems, on-chain attribution, and verifiable work histories. This infrastructure aims to enable global participation in AI training while maintaining data quality standards.
The crypto-native approach allows contributors worldwide to participate in AI development through validated contributions, establishing permanent records of their work on blockchain networks.
Vance Spencer, Co-Founder of Framework Ventures, emphasized the strategic alignment between improved data quality and blockchain incentive structures. “We believe the pace of AI progress will be driven more by better data than by simply scaling models” Spencer stated.
Platform Capabilities
The self-serve platform supports multimodal data collection across audio, image, and video formats. It facilitates reinforcement learning from human feedback (RLHF) and assistant fine-tuning processes throughout the AI development lifecycle.
Perle’s technology enables teams to collect, annotate, and evaluate specialized training data with improved accuracy and speed compared to traditional methods. The platform already supports several companies in scaling their AI training operations across various use cases.
Leadership and Backing
The founding team comprises veterans from Scale AI, Meta, MIT, and Berkeley, bringing extensive experience in AI development and data training methodologies. Their combined expertise spans both technical implementation and practical application of machine learning systems.
Framework Ventures led the current funding round, building on the firm’s track record of early investments in AI, decentralized finance, and blockchain infrastructure projects. The previous $8.5 million seed round in October 2024 was led by CoinFund, with participation from Protagonist, HashKey, and Peer VC.
Market Timing and Vision
The funding arrives as AI companies increasingly recognize data quality as a competitive advantage. Traditional scaling approaches focused primarily on model size and computational power, but industry attention is shifting toward training data sophistication.
Perle’s vision centers on making high-quality human feedback a foundational element of next-generation AI development. The company aims to ensure AI progress becomes not only faster but more reliable and inclusive through decentralized contribution models.
The platform’s launch comes at a time when AI systems require more nuanced training approaches to handle edge cases and complex scenarios that automated systems struggle to interpret correctly.