TL;DR: Brevis is partnering with NeoSoul to explore how our zero-knowledge stack can help keep the instructions traders give AI agents private, starting with NeoTrade, NeoSoul’s self-custodial AI trading agent for prediction markets. The two teams will also work on making NeoSoul’s ecosystem rewards verifiable by anyone.
What NeoSoul Is Building
NeoSoul is building infrastructure for the AI economy across BNB Chain and 0G. Its flagship product, NeoTrade, is an AI trading agent for prediction markets that runs on the user’s own machine, trades within the limits the user sets, and gets better with every market outcome.
NeoSoul is also designing an incentive program that rewards the agent developers and users who take part in its ecosystem.
Your Instructions Are Your Strategy
Most conversations about trader privacy focus on what’s visible on-chain. AI agents add a new place where strategy can leak: the prompt.
An agent’s analysis is only as good as the instructions behind it, and the more precisely a trader spells out a thesis, targets and limits, the better the result. But those instructions, and the analysis that comes back, often reveal exactly what a user plans to trade next, and they pass through third-party AI models along the way. Many traders respond by keeping their questions vague, which protects the strategy but weakens the analysis.
This is the first area Brevis and NeoSoul will explore. NeoTrade will look at combining Pico zkVM with zkTLS, which proves that data came from a specific authenticated source without exposing the data itself, to keep prompts and model responses private. The goal is that only the user and a frontier model provider can see each request and response. If the model runs inside a TEE, a sealed hardware enclave, that visibility could further narrow to the user alone.
Rewards Anyone Can Check
The second area is NeoSoul’s incentive program. NeoSoul wants every reward it announces and distributes to be something the recipient can check against public criteria, rather than take on trust.
This is where zero-knowledge proofs fit naturally: each reward calculation can come with a proof that it followed the published rules, which anyone can check for themselves. Brevis and NeoSoul will work on bringing this to the developers and users building NeoSoul’s agent ecosystem.
Trust, From Prompt to Payout
AI agents are moving from giving advice to taking action, and trading is one of the first places that shift is happening. For it to work, people need to trust what happens to the instructions they hand an agent, and the developers and users growing an ecosystem need to trust how they’re rewarded. These are the two challenges Brevis and NeoSoul aim to address through our collaboration, and we’ll share more as it develops.
About Brevis
Brevis is a verifiable computing platform powered by zero-knowledge proofs, serving as the infinite compute layer for Web3 and beyond. Applications can offload expensive computation off-chain while proving every result on-chain. The Brevis stack includes Pico zkVM for general-purpose computation, which in its 3.1 release proves Ethereum blocks in real time on a single machine with 4 GPUs and 128 bits of provable security, and is fully open source; the ZK Data Coprocessor for trustless access to historical blockchain data; Vera for ZK-proven media authenticity; and ProverNet, the decentralized marketplace for ZK proof generation, live on mainnet. To date, Brevis has generated nearly 400 million proofs across 50+ protocols on 14 blockchains.
Dive Deeper into Brevis:
Website | X | Discord | Pico zkVM | ZK Data Coprocessor | Incentra | ProverNet | Vera | Arca
Interested in building with Brevis? Reach out to us to explore ideas!
