Unforgeable Endorsements Technical Deep-Dive - Nick's Blog

Building Unforgeable Professional Endorsements with ATProtocol - Nick's Blog

Traditional professional endorsements on platforms like LinkedIn lack cryptographic proof—anyone could forge them, and the platform controls the truth. This article introduces a two-record architecture using ATProtocol's Content Identifiers (CIDs) and Decentralized Identifiers (DIDs) to create mathematically unforgeable mutual attestations. By separating proof creation from endorsement acceptance and leveraging the firehose for distributed validation, we build a system where both parties cryptographically consent and no central authority can manipulate the record.

Automated Verification of Proofs in the Universal Composability Framework with Markov Decision Processes

Designing cryptographic protocols and proving these rigorously secure is an arduous and challenging task. Among the methods commonly used to prove security of cryptographic protocols, formalizing it in Canneti's Universal Composability (UC) Framework offers several benefits: (1) Modular design, (2) demonstrating that security remains under arbitrary composition and concurrent execution, (3) the security against any computationally polynomially bound adversary. However, working within the UC Framework can be cumbersome, requires a long time commitment by the prover, and it is prone to errors. While utilization of proof assistants in Cryptography and IT Security is a prominent research area, proof assistants for UC are still in their infancy. Here we show our ongoing work to utilize model checking for verification of proofs in the UC Framework, which to the best of our knowledge is the first attempt to do so. In this work we (1) formally create a Markov Decision Process (MDP) encoding a given proof in the UC Framework, (2) define and proof notions of soundness and completeness for the constructed MDP, (3) implement a proof of concept and (4) demonstrate practical feasibility through experimental evaluation. In summary, in this work we lay out the formal foundations for model checking UC proofs and create a tool that can not only be used for proof verification but also as an assistant for developing proofs in the UC Framework.

badge.blue — CID-First Attestation Specification

Specification for CID-first attestations on AT Protocol records. Inline and remote cryptographic signatures with replay-attack prevention.

Building a Browser-Native Verification Stack for Tinfoil

Learn how we built a browser-based confidentiality and integrity verifier with implementations of browser-native Sigstore and TUF libraries.

ZKPROV: A Zero-Knowledge Approach to Dataset Provenance for Large Language Models

As large language models (LLMs) are used in sensitive fields, accurately verifying their computational provenance without disclosing their training datasets poses a significant challenge, particularly in regulated sectors such as healthcare, which have strict requirements for dataset use. Traditional approaches either incur substantial computational cost to fully verify the entire training process or leak unauthorized information to the verifier. Therefore, we introduce ZKPROV, a novel cryptographic framework allowing users to verify that the LLM's responses to their prompts are trained on datasets certified by the authorities that own them. Additionally, it ensures that the dataset's content is relevant to the users' queries without revealing sensitive information about the datasets or the model parameters. ZKPROV offers a unique balance between privacy and efficiency by binding training datasets, model parameters, and responses, while also attaching zero-knowledge proofs to the responses generated by the LLM to validate these claims. Our experimental results demonstrate sublinear scaling for generating and verifying these proofs, with end-to-end overhead under 3.3 seconds for models up to 8B parameters, presenting a practical solution for real-world applications. We also provide formal security guarantees, proving that our approach preserves dataset confidentiality while ensuring trustworthy dataset provenance.

Backdoor or Feature? A New Perspective on Data Poisoning

In a backdoor attack, an adversary adds maliciously constructed ("backdoor") examples into a training set to make the resulting model vulnerable to manipulation. Defending against such attacks---that is, finding and removing the backdoor examples---typically involves viewing these examples as outliers and using techniques from robust statistics to detect and remove them. In this work, we present a new perspective on backdoor attacks. We argue that without structural information on the training data distribution, backdoor attacks are indistinguishable from naturally-occuring features in the data (and thus impossible to ``detect'' in a general sense). To circumvent this impossibility, we assume that a backdoor attack corresponds to the strongest feature in the training data. Under this assumption---which we make formal---we develop a new framework for detecting backdoor attacks. Our framework naturally gives rise to a corresponding algorithm whose efficacy we show both theoretically and experimentally.

ATProtocol Attestations: Cryptographic Signatures for the Decentralized Web - Nick's Blog

This post introduces the formal ATProtocol attestation specification, a framework for adding cryptographic signatures to ATProto records through two complementary patterns: inline attestations that embed signatures directly in records, and remote attestations that store proof in separate repository records. The specification prevents replay attacks through repository binding, uses CID-based content addressing for integrity, and provides the cryptographic foundation for verified credentials, trusted content, and authenticated interactions in the decentralized ATProtocol ecosystem.

A New Form of Verification on Bluesky - Bluesky

We’re introducing a new layer of verification — a user-friendly, easily recognizable badge. Additionally, independent organizations can verify accounts directly through our Trusted Verifiers feature.

Introducing gpt-oss-safeguard

OpenAI introduces gpt-oss-safeguard—open-weight reasoning models for safety classification that let developers apply and iterate on custom policies.

Building the world's trusted identity platform • Yoti

Our comprehensive suite of customer verification tools make it easy for businesses to be compliant and safe for people to prove who they are.

Membership Inference Attacks From First Principles

A membership inference attack allows an adversary to query a trained machine learning model to predict whether or not a particular example was contained in the model's training dataset. These attacks are currently evaluated using average-case "accuracy" metrics that fail to characterize whether the attack can confidently identify any members of the training set. We argue that attacks should instead be evaluated by computing their true-positive rate at low (e.g.,

SyRA: Sybil-Resilient Anonymous Signatures with Applications to Decentralized Identity

We study Sybil-Resilient Anonymous (SyRA) signatures, a cryptographic primitive that enables credentialed users to generate, on demand, unlinkable pseudonyms tied to any given context, and issue signatures on behalf of these pseudonyms. Concretely, SyRA allows a distributed issuer to turn any legacy identity or personhood identifier, possibly of low entropy, into a unique associated cryptographic key of high pseudoentropy, for use in generating signatures for any given context. Sybil-resilient anonymous signatures achieve three main objectives: 1) Sybil resilience: every user is entitled to at most one digital identity, 2) anonymity: no information about the user’s real identity is leaked, and 3) non-interactive context switching: users can create on their own at most one credential for any given context in a manner that is unlinkable across contexts. We conceptualize the SyRA primitive as an ideal functionality in the Universal Composition (UC) setting and put forth SASSI, an efficient, pairing-based construction that realizes it by utilizing two levels of verifiable random functions (VRFs), a design which may be of independent interest. The first level consists of threshold VRF issuance of a user’s unique secret key tied to their real-world identifier. The second level allows a user to create signatures for each context, under a unique pseudonym per context. Compared to prior cryptographic tools capable of realizing SyRA, SASSI has the unique feature that issuers are stateless and hence do not need to retain any information about past user interactions, a relevant property for a decentralized implementation. We overview various applications of SASSI in multiparty systems, such as cryptocurrency account management and airdrops, e-voting (e.g., for decentralized governance), and privacy-preserving regulatory compliance (e.g., AML/CFT checks). In the context of creating addresses for digital assets, SyRA signatures enable users to embed their legacy identity into their address in a manner that protects their privacy for each application with which they interact. We demonstrate the practicality of SASSI by providing an implementation and performance evaluation of our construction.

Cognitive Integrity Framework: Formal Foundations for Multiagent Security

Multiagent AI systems introduce cognitive attack surfaces absent in single-model inference. When agents delegate to agents, forming beliefs about beliefs through recursive trust hierarchies, manipulation of reason- ing processes—rather than mere data corruption—becomes a primary security concern. This paper presents the Cognitive Integrity Framework (CIF), providing formal foundations for cognitive security in multiagent operators. We develop four interconnected theoretical contributions: a Trust Calculus with bounded delegation (exponential 𝛿𝑑 decay) that prevents trust amplification through delegation chains; a Defense Com- position Algebra with series and parallel composition theorems establishing multiplicative detection bounds; Information-Theoretic Limits relating stealth constraints to maximum attack impact through a fundamental stealth-impact tradeoff; and a formal Adversary Hierarchy (Ω1–Ω5) characterizing external, peripheral, agent-level, coordination, and systemic threats with increasing capability and decreasing detectability. The framework provides complete coverage of the OWASP Top 10 for Agentic Applications through formal threat models grounded in cognitive state manipulation rather than traditional input/output filtering. CIF bridges classical security concepts with the cognitive requirements of agentic systems. We extend Byzantine fault tolerance to cognitive manipulation—agents that appear functional but hold corrupted beliefs—and adapt trust management systems to continuous trust evolution with provable decay bounds. The framework formalizes five architectural defense mechanisms (cognitive firewalls, belief sandboxing, behavioral tripwires, provenance tracking, Byzantine consensus) with composition rules enabling formal reasoning about layered security. Technical foundations include: operational semantics for message passing and trust updates; invariants for belief integrity, goal preservation, and trust boundedness; model checking configurations for safety property verification; and a complete notation system for attack parameterization, defense specification, and cognitive state representation. This is Part 1 of a three-part series: Part 1 (this paper, DOI: 10.5281/zenodo.18364119) presents formal foundations and theoretical analysis; Part 2 (DOI: 10.5281/zenodo.18364128) provides computational validation and implementation; Part 3 (DOI: 10.5281/zenodo.18364130) offers practical deployment guidance. The framework will continue to be developed and versioned at https://github.com/docxology/cognitive_integrity/ .

Targeted Backdoor Attacks on Deep Learning Systems Using Data Poisoning

Deep learning models have achieved high performance on many tasks, and thus have been applied to many security-critical scenarios. For example, deep learning-based face recognition systems have been used to authenticate users to access many security-sensitive applications like payment apps. Such usages of deep learning systems provide the adversaries with sufficient incentives to perform attacks against these systems for their adversarial purposes. In this work, we consider a new type of attacks, called backdoor attacks, where the attacker's goal is to create a backdoor into a learning-based authentication system, so that he can easily circumvent the system by leveraging the backdoor. Specifically, the adversary aims at creating backdoor instances, so that the victim learning system will be misled to classify the backdoor instances as a target label specified by the adversary. In particular, we study backdoor poisoning attacks, which achieve backdoor attacks using poisoning strategies. Different from all existing work, our studied poisoning strategies can apply under a very weak threat model: (1) the adversary has no knowledge of the model and the training set used by the victim system; (2) the attacker is allowed to inject only a small amount of poisoning samples; (3) the backdoor key is hard to notice even by human beings to achieve stealthiness. We conduct evaluation to demonstrate that a backdoor adversary can inject only around 50 poisoning samples, while achieving an attack success rate of above 90%. We are also the first work to show that a data poisoning attack can create physically implementable backdoors without touching the training process. Our work demonstrates that backdoor poisoning attacks pose real threats to a learning system, and thus highlights the importance of further investigation and proposing defense strategies against them.