Shujun Li

Operational technology (OT) devices run safety-critical physical processes, yet their security testing remains manual and expert-intensive. Autonomous security agents could ease this burden, but bare-metal OT devices are difficult targets because they lack shells, filesystems, and named exploit abstractions, requiring operation through protocol fields and observable device state. We present APIOT (Autonomous Purple-teaming for Industrial OT), a large language model (LLM) agent framework that completes a discovery → exploitation → network-level mitigation → verification workflow without human intervention. We evaluate APIOT in 360 runs on vulnerable firmware and simulators for Zephyr RTOS, spanning five LLMs, three topologies, two impairment levels, and prompt, tooling and two ablations. APIOT completed 90% of primary missions (63/70 across protocols and topologies) and 85% of 150 cross-model runs (128/150) — feasible across models, with substantial variation. Named exploit tools (e.g., a canned CoAP option-overflow) were faster, but protocol primitives (the agent builds the fields itself) produced more reusable mitigations that blocked unseen payload variants. Runtime oversight improved reliability: success rose from 27/30 to 29/30 with rule-based guards and 30/30 when a second LLM added advisory steering. To our knowledge, this is the first end-to-end autonomous attack-remediation demonstration on bare-metal industrial devices, establishing controlled feasibility — not zero-day discovery or production-OT transfer.

Machine learning (ML) based classifiers that take a privacy policy as the input and predict relevant concepts are useful in different applications such as (semi-)automated compliance analysis against requirements of a specific data protection law such as the EU GDPR. Although many researchers have studied ML-based privacy policy concept classifiers, we observed multiple research gaps, e.g., the lack of a more complete GDPR taxonomy and the less consideration of hierarchical information in privacy policies. To fill such research gaps, we produced a more complete GDPR-oriented privacy policy concept taxonomy, constructed the first privacy policy corpus with explicitly hierarchical information at three levels, and conducted the most comprehensive performance evaluation study of GDPR concept classifiers for privacy policies, cover many aspects that have not been studied systematically. Our work led to multiple findings and insights, including the usefulness of considering hierarchical contextual features and different hierarchical structures, the observation that a “one size fits all” approach may not work, the reduced performance of such classifiers on our newly constructed corpus especially after the first level, and the necessity to split the training and testing sets by documents.

Privacy policies help inform people about organisations' personal data processing practices, covering different aspects such as data collection, data storage, and sharing of personal data with third parties. Privacy policies are often difficult for people to fully comprehend due to the lengthy and complex legal language used and inconsistent practices across different sectors and organisations. To help conduct automated and large-scale analyses of privacy policies, many researchers have studied applications of machine learning and natural language processing techniques, including large language models (LLMs). While a limited number of prior studies utilised LLMs for extracting personal data flows from privacy policies, our approach builds on this line of work by combining LLMs with retrieval-augmented generation (RAG) and a customised knowledge base derived from existing studies. This paper presents the development of LADFA, an end-to-end computational framework, which can process unstructured text in a given privacy policy, extract personal data flows and construct a personal data flow graph, and conduct analysis of the data flow graph to facilitate insight discovery. The framework consists of a pre-processor, an LLM-based processor, and a data flow post-processor. We demonstrated and validated the effectiveness and accuracy of the proposed approach by conducting a case study that involved examining ten selected privacy policies from the automotive industry. Moreover, it is worth noting that LADFA is designed to be flexible and customisable, making it suitable for a range of text-based analysis tasks beyond privacy policy analysis.

National Computer Security Incident Response Teams (CSIRTs) are established worldwide to coordinate responses to cyber security incidents at the national level. It is known that software tools (including open source ones) and public data are routinely used to facilitate incident response in national CSIRTs. However, there is a lack of an authoritative set of criteria that can be used for a systematic evaluation to decide which software tools and data sources should be used by national CSIRTs for incident response. A prior study identified a set of potential candidate criteria for such an evaluation. The study presented in this article aims to validate these candidate criteria empirically by asking staff members of several national CSIRTs how they perceive the candidate criteria’s practical usefulness and readiness for deployment in national CSIRTs’ operations. The study involved online semi-structured interviews with nine interviewees from nine national CSIRTs in Asia-Pacific, Africa and Europe. After validating the candidate criteria using semi-structured interviews with these nine interviewees, we applied the criteria to evaluate a selection of software tools and data sources by converting each criterion into one or more relevant metrics, such as ‘measuring the time taken by a tool to produce results’. Results from the study led to the following main findings: (1) all interviewees perceived the candidate criteria as practically useful for evaluating tools and data sources in the operations of national CSIRTs; (2) all interviewees agreed that the candidate criteria could be deployed in national CSIRTs and other types of CSIRTs and (3) the candidate criteria can be applied relatively easily in practice. These criteria are envisaged to help national CSIRTs select the most appropriate tools and data sources to facilitate effective incident response, improve their operational practices and improve the quality of wider security operations.