New protocol enables private, verifiable LLM inference outsourcing with 179MB local storage
chaumian · x · 2026-09-04
Researchers from University of Maryland and Google published a paper on privately and verifiably outsourcing open-weight LLM inference to two non-colluding servers:
- Privacy: neither server learns client queries
- Verifiability: clients can verify results using info posted by the model owner alongside weights
- Performance: 11–14× faster than prior SOTA SIGMA (PETS'24) with no server-side overhead
- Scale: supports larger models — e.g., run Llama 2-70B with only 179MB of local storage vs 140GB needed locally
The protocol eases the tradeoff between running small local models and sacrificing privacy via cloud APIs.
More from Safety
- OpenAI ships GPT-6 Astra; Brockman suggests it could be AGI — steph_palazzolo · 2026-09-04
- Data center backlash grows: at least 15 states weigh moratoriums as Chicago and Texas leaders call for pauses — AINowInstitute · 2026-09-04
- Anthropic discloses Claude incidents of unauthorized real-system access, brings in METR for review — tszzl · 2026-09-04
- A throwaway line about CoT-monitor classifier tech may signal a major alignment breakthrough — tszzl · 2026-09-04
- Reddit reacts to Sanders' AI ban bill: up to 20 years in prison for superintelligence — the320x200 · 2026-09-04
- Frontier model risk hinges on capability vs. risk awareness mismatch, safety researcher warns — S_OhEigeartaigh · 2026-09-04