Northeastern's GENESIS AI Develops 5G/6G RAN Features in Hours
Northeastern University's Institute for Intelligent Networked Systems (INSI) developed GENESIS, an agentic AI system that automates the R&D lifecycle for 5G/6G Radio Access Network (RAN) software. This technology can translate specification clauses into validated, over-the-air code in hours, a process that traditionally takes months of manual engineering. GENESIS represents a significant technical development in the application of AI to complex network infrastructure development.
Key Takeaways
- GENESIS converts 3GPP specification clauses or research ideas directly into validated, over-the-air code on production 5G hardware.
- The system autonomously plans, codes, tests, and iterates across simulations, channel emulation, and live over-the-air transmissions on Open6G testbeds.
- In experiments, GENESIS achieved a 100% success rate in implementing 5G features, while a baseline coding agent failed to produce working implementations.
- The framework incorporates agents, skills, and hooks into six autonomous pipelines: SYNTHESIZE, TEST, HARDEN, OPTIMIZE, DISCOVER, and SECURE.
- A shared knowledge plane, SYNAPSE, grounds all agent decisions in 3GPP and O-RAN specifications and accumulates results to compound capabilities.
Additional Context
Northeastern University's Institute for Intelligent Networked Systems (INSI) has been active in advancing AI-native wireless systems. In February 2026, INSI, SoftBank Corp., Keysight Technologies, and zTouch Networks demonstrated AgentRAN at MWC Barcelona 2026 (per INSI, February 2026). This system coordinates a hierarchy of AI agents to translate natural language operator intents into real-time 5G and 6G network configurations, powered by SoftBank's Large Telecom Model (LTM). The demonstration highlighted the system's ability to autonomously adapt to changing network conditions, with performance gains attributed to the LTM and high-fidelity training data. Earlier, in May 2026, Northeastern University's INSI Open6G OTIC also demonstrated the first open-source prototype of a massive MIMO (mMIMO) AI-RAN system (per PR Newswire, May 2026). This prototype integrated O-RAN compliant mMIMO hardware with open-source software and NVIDIA GPU-accelerated RAN intelligence, showcasing sustained throughput for multiple users under mobility. These initiatives, including GENESIS, underscore a consistent push by Northeastern University and its partners to integrate AI and open-source principles into the core of next-generation wireless network development, aiming for greater automation, efficiency, and flexibility in 5G and future 6G deployments.
Read full article at morningstar.com
Get this in your inbox → Subscribe
Enjoy our coverage?
Add StreamingMeme as a preferred source on Google to see more of our streaming news at the top of your Search results.
Add as preferred source