Explanipedia

Comparing AI Agents to Cybersecurity Professionals in Real-World Penetration Testing Open

Justin W. Lin, E. Jones, Donovan Julian Jasper, Ethan Ho, Anna H. Wu , et al. · 2025

We present the first comprehensive evaluation of AI agents against human cybersecurity professionals in a live enterprise environment. We evaluate ten cybersecurity professionals alongside six existing AI agents and ARTEMIS, our new agent …

ZeroOS: A Universal Modular Library OS for zkVMs Open

Guangxian Zou, Isaac Zhang, Ryan Zarick, K.W. Wong, Thomas Kim , et al. · 2025

zkVMs promise general-purpose verifiable computation through ISA-level compatibility with modern programs and toolchains. However, compatibility extends further than just the ISA; modern programs often cannot run or even compile without an…

Context-Dependent Threshold Decryption and Its Applications Open

Dan Boneh, Kartik Nayak, Lior Rotem, Victor Shoup · 2025

BountyBench: Dollar Impact of AI Agent Attackers and Defenders on Real-World Cybersecurity Systems Open

Andy K. Zhang, Joey Ji, Celeste Menders, Riya Dulepet, Thomas Qin , et al. · 2025

AI agents have the potential to significantly alter the cybersecurity landscape. Here, we introduce the first framework to capture offensive and defensive cyber-capabilities in evolving real-world systems. Instantiating this framework with…

ExpProof : Operationalizing Explanations for Confidential Models with ZKPs Open

Chhavi Yadav, Evan Monroe Laufer, Dan Boneh, Kamalika Chaudhuri · 2025

In principle, explanations are intended as a way to increase trust in machine learning models and are often obligated by regulations. However, many circumstances where these are demanded are adversarial in nature, meaning the involved part…

Kite: How to Delegate Voting Power Privately Open

Kamilla Nazirkhanova, Vrushank Gunjur, X. Pilli Cruz-De Jesus, Dan Boneh · 2025

Ensuring the privacy of votes in an election is crucial for the integrity of a democratic process. Often, voting power is delegated to representatives (e.g., in congress) who subsequently vote on behalf of voters on specific issues. This d…

zkPi: Proving Lean Theorems in Zero-Knowledge Open

Evan Laufer, Alex Ozdemir, Dan Boneh · 2024

Cryptography and Computer Security: A View From the Year 2100 Open

Dan Boneh · 2024

Cybench: A Framework for Evaluating Cybersecurity Capabilities and Risks of Language Models Open

Andy K. Zhang, Neil Perry, Riya Dulepet, Eliot Jones, Justin W. Lin , et al. · 2024

Language Model (LM) agents for cybersecurity that are capable of autonomously identifying vulnerabilities and executing exploits have potential to cause real-world impact. Policymakers, model providers, and researchers in the AI and cybers…

Divisible E-Cash for Billing in Private Ad Retargeting Open

Kevin Liao, Henry Corrigan-Gibbs, Dan Boneh · 2024

This paper presents new techniques for private billing in systems for privacy-preserving online advertising. In particular, we show how an ad exchange can use an e-cash scheme to bill advertisers for ad impressions without learning which c…

A Survey of Two Verifiable Delay Functions Using Proof of Exponentiation Open

Dan Boneh, Benedikt Bünz, Ben Fisch · 2024

A verifiable delay function (VDF) is an important tool used for adding delay in decentralized applications. This paper surveys and compares two beautiful verifiable delay functions, one due to Pietrzak, and the other due to Wesolowski, In …

Optimistic Verifiable Training by Controlling Hardware Nondeterminism Open

Megha Srivastava, Simran Arora, Dan Boneh · 2024

The increasing compute demands of AI systems have led to the emergence of services that train models on behalf of clients lacking necessary resources. However, ensuring correctness of training and guarding against potential training-time a…

FairProof : Confidential and Certifiable Fairness for Neural Networks Open

Chhavi Yadav, Amrita Roy Chowdhury, Dan Boneh, Kamalika Chaudhuri · 2024

Machine learning models are increasingly used in societal applications, yet legal and privacy concerns demand that they very often be kept confidential. Consequently, there is a growing distrust about the fairness properties of these model…

R-Pool and Settlement Markets for Recoverable ERC-20R Tokens Open

Kaili Wang, Qinchen Wang, C. S. Cai, Dan Boneh · 2023

ERC-20R is a wrapper around ERC-20 that supports asset recovery within a limited time window after an asset is transferred. It is designed to reduce theft and losses on the blockchain by allowing a victim to recover their stolen or lost as…

Open Problems in DAOs Open

Joshua Tan, Tara Merk, Sarah Hubbard, Eliza R. Oak, Ellie Rennie , et al. · 2023

Decentralized autonomous organizations (DAOs) are a new, rapidly-growing class of organizations governed by smart contracts. Here we describe how researchers can contribute to the emerging science of DAOs and other digitally-constituted or…

Vector Commitments with Efficient Updates Open

Ertem Nusret Tas, Dan Boneh · 2023

Dynamic vector commitments that enable local updates of opening proofs have applications ranging from verifiable databases with membership changes to stateless clients on blockchains. In these applications, each user maintains a relevant s…

Do Users Write More Insecure Code with AI Assistants? Open

Neil Perry, Megha Srivastava, Deepak Kumar, Dan Boneh · 2022

We conduct the first large-scale user study examining how users interact with an AI Code assistant to solve a variety of security related tasks across different programming languages. Overall, we find that participants who had access to an…

zkBridge Open

Tiancheng Xie, Jiaheng Zhang, Zerui Cheng, Fan Zhang, Yupeng Zhang , et al. · 2022

Blockchains have seen growing traction with cryptocurrencies reaching a market cap of over 1 trillion dollars, major institution investors taking interests, and global impacts on governments, businesses, and individuals.

zkBridge: Trustless Cross-chain Bridges Made Practical Open

Tiancheng Xie, Jiaheng Zhang, Zerui Cheng, Fan Zhang, Yupeng Zhang , et al. · 2022

Blockchains have seen growing traction with cryptocurrencies reaching a market cap of over 1 trillion dollars, major institution investors taking interests, and global impacts on governments, businesses, and individuals. Also growing signi…

Memory Tagging: A Memory Efficient Design Open

Aditi Partap, Dan Boneh · 2022

ARM recently introduced a security feature called Memory Tagging Extension or MTE, which is designed to defend against common memory safety vulnerabilities, such as buffer overflow and use after free. In this paper, we examine three aspect…

Cryptoeconomic Security for Data Availability Committees Open

Ertem Nusret Tas, Dan Boneh · 2022

Layer 2 systems have received increasing attention due to their potential to scale the throughput of L1 blockchains. To avoid the cost of putting data on chain, these systems increasingly turn to off-chain data availability solutions such …

ERC-20R and ERC-721R: Reversible Transactions on Ethereum Open

Kaili Wang, Qinchen Wang, Dan Boneh · 2022

Blockchains are meant to be persistent: posted transactions are immutable and cannot be changed. When a theft takes place, there are limited options for reversing the disputed transaction, and this has led to significant losses in the bloc…

Strong Anonymity for Mesh Messaging Open

Neil Perry, Bruce Spang, Saba Eskandarian, Dan Boneh · 2022

Messaging systems built on mesh networks consisting of smartphones communicating over Bluetooth have been used by protesters around the world after governments have disrupted Internet connectivity. Unfortunately, existing systems have been…

Clarion: Anonymous Communication from Multiparty Shuffling Protocols Open

Saba Eskandarian, Dan Boneh · 2022

This paper studies the role of multiparty shuffling protocols in enabling more efficient metadatahiding communication.We show that the process of shuffling messages can be expedited by having servers collaboratively shuffle and verify secr…

Secure Complaint-Enabled Source-Tracking for Encrypted Messaging Open

Charlotte Peale, Saba Eskandarian, Dan Boneh · 2021

While the end-to-end encryption properties of popular messaging schemes such as Whatsapp, Messenger, and Signal guarantee privacy for users, these properties also make it very difficult for messaging platforms to enforce any sort of conten…

Attacks on Onion Discovery and Remedies via Self-Authenticating Traditional Addresses Open

Paul Syverson, Matthew Finkel, Saba Eskandarian, Dan Boneh · 2021

Onion addresses encode their own public key. They are thus\nself-authenticating, one of the security and privacy advantages of onion\nservices, which are typically accessed via Tor Browser. Because of the mostly\nrandom-looking appearance …

Empirical Measurements on Pricing Oracles and Decentralized Governance for Stablecoins Open

Wanyun Catherine Gu, Anika Raghuvanshi, Dan Boneh · 2021

Stablecoins are designed to address the volatility of crypto assets by maintaining a peg to a non-volatile currency such as the US Dollar.Decentralized Stablecoins that maintain their collateral on-chain need a pricing oracle to determine …

Lightweight Techniques for Private Heavy Hitters Open

Dan Boneh, Elette Boyle, Henry Corrigan-Gibbs, Niv Gilboa, Yuval Ishai · 2021

This paper presents Poplar, a new system for solving the private heavy-hitters problem. In this problem, there are many clients and a small set of data-collection servers. Each client holds a private bitstring. The servers want to recover …

Supersingular curves with small noninteger endomorphisms Open

Jonathan Love, Dan Boneh · 2020

We introduce a special class of supersingular curves over ‫ކ‬ p 2 , characterized by the existence of noninteger endomorphisms of small degree.We prove a number of properties about this set.Most notably, we can partition this set into subs…

Differentially Private Learning Needs Better Features (or Much More Data) Open

Florian Tramèr, Dan Boneh · 2020

We demonstrate that differentially private machine learning has not yet reached its "AlexNet moment" on many canonical vision tasks: linear models trained on handcrafted features significantly outperform end-to-end deep neural networks for…

Dan Boneh YOU? Author Swipe