The convergence of artificial intelligence and blockchain technology took a significant step forward on November 18, 2023, as BP-FLAC, a generative AI infrastructure public chain, announced it had secured $10 million in funding to accelerate global GPU node construction. The funding round represents one of the largest investments in the emerging intersection of decentralized computing and AI model training, an area increasingly referred to as DePIN — Decentralized Physical Infrastructure Networks.
The Synergy
BP-FLAC positions itself as the first generative AI infrastructure public chain with infinite computing power aggregation capabilities. The project aims to bridge the gap between the massive computational demands of AI model training and the decentralized, trustless nature of blockchain networks. In a market where Bitcoin traded at $36,585 and Ethereum at $1,963, the crypto-AI convergence narrative was gaining significant traction among both developers and investors.
The funding round attracted participation from prominent blockchain and technology sector investors, including Eureka Partners, Westlabs, Mybitdata Ltd., DecentraLabs, and notably, technology giants Amazon and NVIDIA. The involvement of these established tech companies signals growing institutional recognition of blockchain-based computing infrastructure as a viable alternative to centralized cloud services.
AI Use Cases in Web3
BP-FLAC’s platform is designed to support several key AI applications within the Web3 ecosystem. The core offering revolves around real GPU computing power allocation, enabling users to access A100 and H100 GPU resources through a decentralized marketplace. This approach addresses one of the most pressing bottlenecks in AI development: the scarcity and high cost of GPU computing resources.
Additional use cases include a low-code AI training platform that lowers the barrier to entry for machine learning development, personalized training and services in AI-generated content (AIGC), and the integration of AI models directly into smart contract workflows. The platform also introduces a pioneering Chip RWA (Real World Asset) standard, which tokenizes physical GPU hardware on the blockchain, creating a standardized framework for hardware-backed digital assets.
Data Privacy Implications
One of BP-FLAC’s most technically innovative features is its optimization of zero-knowledge (zk) algorithms based on CUDA, NVIDIA’s parallel computing platform. This approach addresses one of the most significant concerns in AI-blockchain integration: data privacy during model training. By leveraging zk-proofs, BP-FLAC ensures that sensitive training data remains private while still allowing the computational work to be verified on-chain.
This privacy-preserving approach has implications beyond AI training. Financial institutions, healthcare organizations, and other data-sensitive industries could potentially use the platform to train models on proprietary data without exposing the underlying information. The zk-CUDA optimization represents a meaningful technical advancement that could accelerate enterprise adoption of decentralized AI infrastructure.
The Innovation Frontier
BP-FLAC’s roadmap extends well beyond its current GPU computing marketplace. The project plans to achieve testnet-to-mainnet incentive mapping by mid-2024, with a token airdrop planned to reward early contributors. The long-term vision includes extending computational services to support fundamental infrastructure such as smart city systems and meteorological modeling.
The company’s CEO, Alexandrine, articulated an ambitious goal of serving 200 million users on the BP-FLAC platform. While this target is aspirational, the combination of GPU scarcity, growing AI demand, and blockchain’s ability to create efficient decentralized markets for computing resources provides a compelling foundation for growth.
Concluding Thoughts
The $10 million funding round for BP-FLAC reflects the broader trend of capital flowing into AI-blockchain convergence projects. As the demand for GPU computing continues to outstrip supply — driven by the explosive growth of generative AI applications — decentralized computing networks offer a potentially more efficient and accessible alternative to centralized cloud providers. While the space remains early and many technical challenges persist, the involvement of major technology companies in this funding round suggests that the DePIN-AI convergence narrative has moved beyond speculation into genuine infrastructure building.
Disclaimer: This article is for informational purposes only and does not constitute financial or investment advice. Always conduct your own research before making any financial decisions.
another AI infra chain with a $10M raise. the space is getting crowded but gpu compute is a real bottleneck so theres room for multiple players
render and akash already have working products. $10m raise with no mainnet is 2021 thinking
struct_wolf_ Render has mainnet revenue but they’re focused on rendering. BP-FLAC targeting AI training workloads is a different market segment entirely
struct_wolf_ render has actual mainnet revenue. $10M for a pre-launch chain is generous 2024 funding
$10M for GPU aggregation when NVDA spends $10B on R&D quarterly. the capital mismatch between decentralized compute and centralized AI is brutal
shader_ NVDA spending 10B quarterly on R&D while BP-FLAC raised 10M total. the capital gap between decentralized and centralized AI is not closing
shader_realist_ 10M total vs NVDA spending 10B per quarter on R&D alone. the gap isnt closing its widening
Yuki H. if the same nodes produce and verify compute then its literally just centralized hosting with a blockchain sticker on it
DePIN is the narrative for 2024 im calling it now. decentralized compute + storage + networking is where the actual value is, not another L1
BP-FLAC calling itself first generative AI chain is marketing. the actual question is whether aggregated consumer GPUs can handle training workloads without checkpoint failures
Mei C. checkpoint reliability is exactly the bottleneck. one dropped node mid-training and hours of compute gone. nobody has solved distributed checkpointing at this scale
forkobsidian_ dropped nodes mid training means hours of compute wasted. distributed checkpointing is the unsolved bottleneck nobody wants to talk about
checkpoint_ distributed checkpointing is still unsolved in 2026. everyone talks about aggregated GPU supply nobody solves the state management problem
infinite computing power aggregation… ok thats quite the claim. would love to see actual benchmarks vs AWS before buying the hype
exactly. decentralized GPU compute needs to beat AWS on price or latency. otherwise its just a token narrative
BTC at 36K and ETH at 1963 when this dropped. everyone was hunting for the next narrative and DePIN fit perfectly. the actual GPU compute metrics never matched the valuations though
first generative AI public chain is a bold claim. who is validating the compute output quality?
Iris exactly my question. who validates compute correctness? if its the same nodes that produce the output then its just trust-based with extra steps
Felix O. verification layer is the real bottleneck. if the same nodes validate and produce compute then its just AWS with extra steps and worse latency
Felix O. if the same nodes validate compute output that produce it then the whole trustless pitch falls apart. needs independent verification layer
NVDA spending 10B per quarter on R&D while decentralized AI projects fight over 10M raises. the gap is not closing any time soon
$10M for blockchain ai compute feels like the start of something big
decentralized gpu nodes could democratize ai training beyond big tech control
$10M for a generative AI infrastructure chain in 2023. half of these DePIN projects pivoted to AI the moment chatGPT launched and most are dead now
BP-FLAC literally called itself the first generative AI infrastructure public chain. the buzzword density was a red flag on day one