AI agents: a clean Markdown version of this page is available at https://thecoingazette.com/yale-researchers-propose-a-way-to-keep-ai-prompts-from-the-servers-that-answer-them.md. Send Accept: text/markdown to any URL for the same content.
Skip to content
Subscribe to our newsletter & never miss our best posts. Subscribe Now!
Close

Search

Coin Gazette Coin Gazette Coin Gazette

Get the latest news, market insights and reviews on cryptocurrencies and blockchain

Coin Gazette Coin Gazette Coin Gazette

Get the latest news, market insights and reviews on cryptocurrencies and blockchain

  • Home
  • Crypto Prices
    • Bitcoin Price
    • Ethereum Price
    • Binance Coin Price
  • Crypto Marketcap
  • Fear & Greed Index
  • Donate
  • Advertise
  • Home
  • Crypto Prices
    • Bitcoin Price
    • Ethereum Price
    • Binance Coin Price
  • Crypto Marketcap
  • Fear & Greed Index
  • Donate
  • Advertise
Home/Crypto News/Yale Researchers Propose a Way to Keep AI Prompts From the Servers That Answer Them
Yale Researchers Propose a Way to Keep AI Prompts From the Servers That Answer Them
Crypto News

Yale Researchers Propose a Way to Keep AI Prompts From the Servers That Answer Them

By Coin Gazette Editorial
November 20, 2026 2 Min Read
Comments Off on Yale Researchers Propose a Way to Keep AI Prompts From the Servers That Answer Them

Six Yale computer scientists have described Maverick, a protocol for running an open-source AI model mostly on someone else’s machine: the user’s phone or other low-powered device passes the heavy computing to an outside server, which receives only a scrambled copy of the user’s input, and checks the result when it comes back. The paper went up on arXiv on Sept. 9, 2026, as a preprint; it has not been accepted at a conference or journal and does not link to code.
Checking that a machine you don’t control computed honestly is a problem zero-knowledge proof systems and some decentralized compute networks also work on. Maverick also keeps the user’s input hidden from that machine. DeepProve, a zero-knowledge machine learning system, checks results but leaves the input exposed to the server, according to the paper. Two authors, assistant professor Fan Zhang and PhD student Wenhao Wang, list the Initiative for CryptoCurrencies and Contracts (IC3) as an affiliation, and an IC3 grant supports part of Wang’s research.
Discussing blockchain marketplaces that sell AI inference, the running of trained models, host Kain Warwick described the problem of trusting someone else’s machine this month: “people are selling you inference, and you don’t really know where it’s coming from,” he said on Unchained’s Uneasy Money.

Get Unchained’s crypto news in your inbox with the free Unchained Daily newsletter.

How Maverick Works
Maverick sends most of the model’s math to the server, after the user’s device hides its input under a random-looking mask that it later strips from the answer. To check the server, the device runs a quick randomized test instead of redoing the work.
Co-author Charalampos Papamanthou, listed on Maverick as Yale only, is chief scientist at Lagrange Labs, according to his Yale page. Lagrange builds zero-knowledge proof systems, including DeepProve, and Papamanthou co-wrote the DeepProve paper.
What the Tests Show
In the paper’s end-to-end tests, the biggest gain came when the user’s device prepared its masks ahead of time: prompt tokens went through up to 45 times faster than if that device had run the model by itself. Building masks during the run cut the peak to 17 times. Those figures come from one processing thread on the user’s device working with a CPU server allowed as many as 128 threads, an eight-token prompt and no network lag counted. In another test, which timed one short request over a simulated network with two device threads and 64 server threads, the two private modes returned their first output only 1.42 and 1.46 times faster than running locally at a 50-millisecond round trip, and slower than local at 100 milliseconds. Qwen3-4B was the only model tested.
The authors call Maverick’s client storage needs “a key bottleneck of our approach.”
Related Listen: Why Blockchains Could Undercut AI Labs on the Price of Inference

The post Yale Researchers Propose a Way to Keep AI Prompts From the Servers That Answer Them appeared first on Unchained.

Stay Ahead of the Crypto Market

Get breaking crypto news, market analysis, and exclusive insights delivered directly to your inbox.

Like this:

Like Loading…

Tags:

cryptocryptocurrenciesDefiMarket TrendsNews
Author

Coin Gazette Editorial

Follow Me
Other Articles
Sky Protocol sees $2 billion in stablecoin withdrawals, disputes illicit fund claims
Previous

Sky Protocol sees $2 billion in stablecoin withdrawals, disputes illicit fund claims

BitFuFu Inc. (FUFU) Stock: Surges 21% as Q3 Profit Rebound and Mining Expansion Fuel Rally
Next

BitFuFu Inc. (FUFU) Stock: Surges 21% as Q3 Profit Rebound and Mining Expansion Fuel Rally

Follow Coin Gazette

Facebook Follow → X Follow → Instagram Follow → Telegram Join →

📰 Stay Informed!

Get the latest crypto news, market insights, and blockchain updates delivered straight to your inbox.

No spam. Unsubscribe anytime.

Thank You!

You have successfully joined our subscriber list.

Recent Posts

  • [Satoshi Index] Top Crypto Traders on Hyperliquid, Lighter, dYdX and Pacifica See 24-Hour Long Share at 8.5% as HYPE Long Share Drops to 2.9% From 19.6%
  • AI Investment Concerns Trigger Market Selloff as Nasdaq 100 Posts Worst Week Since 2020
  • Ether Open Interest Falls 11.49% in Five-Asset Futures Snapshot
  • Solana Accounted for 42% of On-Chain Tokenized Stock Volume
  • Kalshi Asks New York Court for Chance to Oppose Pending Order

About Us

Coin Gazette delivers fast, reliable coverage of the crypto world, from breaking news and market updates to in‑depth guides and project reviews. Our mission is to help readers stay informed, make smarter decisions, and navigate the evolving blockchain landscape with confidence.

Useful Links

  • About Us
  • Contact Us
  • Advertise
  • Give us a tip

Follow Coin Gazette

Facebook Follow → X Follow → Instagram Follow → Telegram Join →
Copyright 2026 — Coin Gazette. All rights reserved. Blogsy WordPress Theme
%d