Expert Designed RISC-V SoC Using Meta AI Agent
A semiconductor developer utilized Meta's Muse AI to create a specialized cryptocurrency chip for roughly $2,000.
Updated on Oct. 14, 2026 in Semiconductors
Live Poll
Do you trust AI agents to handle complex engineering design tasks currently performed by humans?
A semiconductor expert has successfully designed a crypto-specific system-on-a-chip (SoC) on the RISC-V architecture using Meta's Muse AI agent. The process, which cost approximately $2,000, marks a significant application of generative AI in hardware engineering.
Why it matters
This design project demonstrates the potential for AI to drastically lower the financial barrier to entry for custom silicon development. By automating complex testing and design tasks, such workflows may accelerate hardware innovation for specialized blockchain applications.
The AI agent consumed 100 million tokens to generate the design and verify it against 1.5 million Berkeley TestFloat vectors. The system passed official ACT suites and represents a move toward native blockchain hardware integration.
The players
Meta
Meta is a multinational technology conglomerate that develops generative AI models including the Muse agent.
NVIDIA
NVIDIA is a technology company that co-launched the Build-A-Chip-Bench benchmark to evaluate AI performance in hardware design.
Anthropic
Anthropic is an AI research firm that developed the Claude Fable 5.1 model.
The details
Meta's Muse AI agent handled the architecture design, software testing, and manufacturing-prep requirements for the chip. The developer plans to refine the architecture by adding custom vector instructions and replacing the initial third-party baseline core.
Timeline
October 14, 2026: The RISC-V SoC design project was published.
The Tech Race
This project fits into the rapidly evolving field of AI-assisted chip design, currently benchmarked against metrics like the Build-A-Chip-Bench. It signals a move away from manual hardware layout toward automated, token-driven silicon generation.
For developers and tech enthusiasts, this indicates that creating custom silicon could soon become significantly more affordable and accessible. The project showcases how AI agents can lower design costs, potentially leading to more specialized hardware for decentralized networks.
The takeaway
The successful creation of a custom RISC-V chip for $2,000 highlights how AI is narrowing the gap between software and hardware engineering. Developers interested in this space should monitor the performance of AI models on the Build-A-Chip-Bench to gauge future capabilities in silicon design.
Further reading
For more information on the evolving landscape of chip development, visit our Semiconductors section.
Source note: This article includes information reported by Wccftech.
Live Poll
Do you trust AI agents to handle complex engineering design tasks currently performed by humans?
