Usual stablecoin issuer just launched its USD0/USDC liquidity pool on Fluid DeFi protocol, allowing liquidity providers to earn dual yields from both lending and trading APRs.
On May 19, an RWA-backed stablecoin protocol Usual announced the launch of its USD0/USDC liquidity pool on the DeFi protocol Fluid. The integration allows liquidity providers to earn lending APR, trading APR, and USUAL rewards on top.
The launch is powered by Fluid’s advanced architecture, which optimizes liquidity ranges and enables deeper, more efficient markets for stablecoin trading. This results in tighter spreads and better execution for users interacting with the USD0/USDC pair.
However, the real edge of USD0 being on Fluid is its relending mechanism, which allows deposited liquidity to simultaneously earn returns from both trading activity and lending protocols — enabling LPs to enjoy dual yield from a single position.
USD0 is a permissionless stablecoin backed by real-world assets — primarily ultra-short maturity U.S. Treasury Bills. It was launched by Usual Protocol to offer greater safety than USD Coin (USDC) and Tether (USDT) by avoiding reliance on traditional banks and their fractional reserve practices. It provides full transparency of its collateral, enabling anyone to verify its backing in real time.
Usual Protocol is headed by CEO Pierre Person, a former French politician and National Assembly member who played a key role in shaping the country’s crypto asset legislation.
The problem: Crypto’s stumbles and AI’s peril
Blockchains and cryptocurrency promised to disrupt industries by eliminating middlemen and streamlining systems like finance and supply chains. Bitcoin 
btc2.11%Bitcoin and stablecoins have found traction, but smart contracts, once revolutionary, have fueled mostly speculative DeFi projects and meme coins rather than real-world solutions. The gap between crypto’s ambition and reality has eroded trust.
AI could end up reshaping everything—from healthcare and science to the way we govern society. But when just a few companies control that kind of power, there’s a real risk of deepening inequality, increasing surveillance, and even steering public opinion. If you look back, technologies like the internet or nuclear energy were developed with heavy government involvement. That’s not the case with AI. It’s largely in the hands of private corporations now, and that brings up a pressing question: Is this technology being built for the common good, or just for profit? Without intervention, AI could follow social media’s path, exploiting users rather than empowering them.
Why decentralization is essential for AI
The breakthrough here is not just technical, but also economic. In decentralized AI networks, every layer of the AI value chain can be distributed in real time. Data custodians who supply datasets, model architects who publish improved weights, and application builders who deliver user experiences can all earn a proportional share of on‑chain rewards. Because every transaction settles on a public blockchain, everyone can audit who earned what and why, creating radical accountability that proprietary labs cannot match.
This structure unlocks a level of collaborative and competitive velocity impossible inside a single company. Thousands of independent nodes iterate in parallel, stress‑testing and improving upon one another’s ideas and forking the best into new sub‑networks. Breakthroughs, therefore, compound rapidly instead of waiting for a quarterly roadmap.
In short, decentralization rewires AI’s incentives so that rewards and governance flow to the true value creators rather than bottling up inside a single balance sheet. That alignment is the difference between an AI future owned by a handful of companies and one that belongs to all of us.
Decentralized AI in action
Bittensor is one of the examples of decentralized AI solutions. Bittensor is a live, open network where crypto-economic incentives translate directly into better AI. Independent nodes post tasks, share weights, and benchmark one another’s output. Every interaction is logged on-chain, and contributors are paid in native token Bittensor 
tao6.81%Bittensor or subnet tokens the moment their work moves the frontier forward.
BitMind, in this economic flywheel, plays the role of a deepfake detector. A swarm of computer‑vision models hunts manipulated images and video. Each week, peer nodes re‑score one another, and detectors that outperform earn larger rewards. The result is an 88 % detection rate, nearly twenty points higher than leading proprietary tools, and real‑time adaptation when new deepfake techniques appear. Moreover, instead of one lab dictating what a language model should be, Templar, a decentralized model training, lets anyone supply data, compute, or architectures to optimize training loss. The subnets’ validators determine algorithmically which contributions improve performance, and rewards flow accordingly.
What binds these projects is the same incentive loop: every incremental improvement, whether a cleaner dataset, an improved model, or improved performance, earns its contributor a larger share of emissions. Open‑source altruism finally has a sustainable business model.
Crypto promised to democratize money but got lost in speculation. Decentralized AI redeems this vision by creating a sustainable incentive and economic model for open-source AI development. If large-scale generalized intelligence will shape the next century, ensuring its rewards are broadly shared may become crypto’s most important, and most achievable legacy.