arXiv subjects

Tom Reed

Publications and source records attributed to Tom Reed.

9 recordsLinked to original sources

arXiv2026

Decoding Decision Correctness from EEG Under High Cognitive Workload in Virtual Reality: Implications for Collaborative Brain-Computer Interface Teams

Collaborative Brain-Computer Interfaces (cBCIs) offer a promising mechanism to augment team decision-making, but existing approaches rely exclusively on evidence available only after a decision has been made and reported, such as reaction time or stated confidence. This limits their use to explaining or discounting a decision after the fact, rather than informing a team's response before it is finalised. We tested whether spatial-covariance EEG features could instead provide a genuinely pre-emptive signal of an operator's decision correctness, available within the response window itself, and whether such a signal depends on cognitive workload. Using a continuous virtual reality target-detection task, participants (N = 23) completed a within-subject workload manipulation (High vs. Low). At the team level, weighting votes by this pre-emptive neural signal, available before a response is committed, produced substantial accuracy gains on contested (evenly-split) trials under High Workload (57% to 88% as team size increased from 2 to 16), but was actively detrimental under Low Workload. Critically, this advantage held even against post-hoc behavioural signals: confidence was the strongest single team-level signal overall, but by definition cannot inform a decision still in progress, whereas the neural signal can. These findings indicate that EEG-based decision-reliability signals are not a general-purpose team augmentation tool, but a workload-conditional one, with clear implications for when and how cBCI systems should be deployed in operational teams.

cs.HC↗

arXiv2026

The Timing Dependencies of Trust: Speed, Accuracy, and cBCI Neuro-Decoupling in Human-AI Teams

The speed and accuracy of an artificial teammate fundamentally alter the failure states of Human-AI integration. While high-speed AI interventions risk inducing reflexive blind compliance, delayed interventions can induce ambiguous cognitive conflict. This study investigates how the fundamental characteristics of an in-task AI assistant, Fast/Less-Accurate (FLA-AI) versus Slow/Accurate (SA-AI) impact the synergy of Collaborative Brain-Computer Interface (cBCI) teams in a Virtual Reality drone task. Seventeen operators completed continuous search tasks under high cognitive workload while their spatial covariance was mapped using a 2D Adaptive Riemannian Oracle. The results mathematically demonstrate that AI timing dictates the mechanism of team failure. Fast AI induced instant, blind compliance; human accuracy under deception collapsed to 50.2%, and pure behavioural teams (N=8) failed to scale beyond 74.1%. In contrast, Slow AI induced delayed cognitive conflict; humans hesitated (61.1% accuracy), but N=8 behavioural teams eventually recovered to 100.0%. Crucially, the Riemannian Oracle mathematically adapted to these states: it heavily restricted temporal windows (< 0.8s) to intercept fast reflexive compliance, while widening windows (> 1.2s) to capture delayed cognitive conflict. Integrating these isolated veridical signals via Hybrid Fusion successfully rescued the Fast AI team (+7.6% at N=8) and significantly accelerated the recovery of smaller Slow AI teams (+6.9% at N=4). These findings prove that cBCI synergy is heavily contingent on the temporal dynamics of trust, providing a critical framework for designing dynamically gated Human-AI systems.

cs.HC↗

arXiv2025

Human-AI Teaming Under Deception: An Implicit BCI Safeguards Drone Team Performance in Virtual Reality

Human-AI teams can be vulnerable to catastrophic failure when feedback from the AI is incorrect, especially under high cognitive workload. Traditional team aggregation methods, such as voting, are susceptible to these AI errors, which can actively bias the behaviour of each individual and inflate the likelihood of an erroneous group decision. We hypothesised that a collaborative Brain-Computer Interface (cBCI) using neural activity collected before a behavioural decision is made can provide a source of information that is "decoupled" from this biased behaviour, thereby protecting the team from the deleterious influence of AI error. We tested this in a VR drone surveillance task where teams of operators faced high workload and systematically misleading AI cues. Using a passive BCI (pBCI) framework validated via offline simulation, we compared traditional behaviour-based team strategies against a purely Neuro-Decoupled Team (NDT) that used only BCI confidence scores derived from pre-response EEG. Under AI deception, behaviour-based teams catastrophically failed, with Majority Vote accuracy collapsing to 42.6% (worse than chance). The NDT, however, maintained a robust 68.3% accuracy. While this did not exceed the best individual's theoretical maximum, it provided a critical +25.7% "Safety Net Delta" that prevented the team from succumbing to the correlated error. This resilience was explained by a neuro-behavioural decoupling, where the BCI's predictions relied on preserved posterior-visual processing ("The Truth Signal") while the operators' executive monitoring systems collapsed. We conclude that an implicit BCI provides resilience by learning to bypass a compromised executive networks and access the preserved sensory representation of ground truth, defending against AI-induced error in high-stakes environments.

cs.HC↗

arXiv2025

What do model reports say about their ChemBio benchmark evaluations? Comparing recent releases to the STREAM framework

Most frontier AI developers publicly document their safety evaluations of new AI models in model reports, including testing for chemical and biological (ChemBio) misuse risks. This practice provides a window into the methodology of these evaluations, helping to build public trust in AI systems, and enabling third party review in the still-emerging science of AI evaluation. But what aspects of evaluation methodology do developers currently include -- or omit -- in their reports? This paper examines three frontier AI model reports published in spring 2025 with among the most detailed documentation: OpenAI's o3, Anthropic's Claude 4, and Google DeepMind's Gemini 2.5 Pro. We compare these using the STREAM (v1) standard for reporting ChemBio benchmark evaluations. Each model report included some useful details that the others did not, and all model reports were found to have areas for development, suggesting that developers could benefit from adopting one another's best reporting practices. We identified several items where reporting was less well-developed across all model reports, such as providing examples of test material, and including a detailed list of elicitation conditions. Overall, we recommend that AI developers continue to strengthen the emerging science of evaluation by working towards greater transparency in areas where reporting currently remains limited.

cs.CY↗

arXiv2025

STREAM (ChemBio): A Standard for Transparently Reporting Evaluations in AI Model Reports

Evaluations of dangerous AI capabilities are important for managing catastrophic risks. Public transparency into these evaluations - including what they test, how they are conducted, and how their results inform decisions - is crucial for building trust in AI development. We propose STREAM (A Standard for Transparently Reporting Evaluations in AI Model Reports), a standard to improve how model reports disclose evaluation results, initially focusing on chemical and biological (ChemBio) benchmarks. Developed in consultation with 23 experts across government, civil society, academia, and frontier AI companies, this standard is designed to (1) be a practical resource to help AI developers present evaluation results more clearly, and (2) help third parties identify whether model reports provide sufficient detail to assess the rigor of the ChemBio evaluations. We concretely demonstrate our proposed best practices with "gold standard" examples, and also provide a three-page reporting template to enable AI developers to implement our recommendations more easily.

cs.CY↗

arXiv2025

Assessing the performance of compartmental and renewal models for learning $R_{t}$ using spatially heterogeneous epidemic simulations on real geographies

The time-varying reproduction number ($R_t$) gives an indication of the trajectory of an infectious disease outbreak. Commonly used frameworks for inferring $R_t$ from epidemiological time series include those based on compartmental models (such as the SEIR model) and renewal equation models. These inference methods are usually validated using synthetic data generated from a simple model, often from the same class of model as the inference framework. However, in a real outbreak the transmission processes, and thus the infection data collected, are much more complex. The performance of common $R_t$ inference methods on data with similar complexity to real world scenarios has been subject to less comprehensive validation. We therefore propose evaluating these inference methods on outbreak data generated from a sophisticated, geographically accurate agent-based model. We illustrate this proposed method by generating synthetic data for two outbreaks in Northern Ireland: one with minimal spatial heterogeneity, and one with additional heterogeneity. We find that the simple SEIR model struggles with the greater heterogeneity, while the renewal equation model demonstrates greater robustness to spatial heterogeneity, though is sensitive to the accuracy of the generation time distribution used in inference. Our approach represents a principled way to benchmark epidemiological inference tools and is built upon an open-source software platform for reproducible epidemic simulation and inference.

q-bio.PE↗

arXiv2024

AI Safety Frameworks Should Include Procedures for Model Access Decisions

The downstream use cases, benefits, and risks of AI models depend significantly on what sort of access is provided to the model, and who it is provided to. Though existing safety frameworks and AI developer usage policies recognise that the risk posed by a given model depends on the level of access provided to a given audience, the procedures they use to make decisions about model access are ad hoc, opaque, and lacking in empirical substantiation. This paper consequently proposes that frontier AI companies build on existing safety frameworks by outlining transparent procedures for making decisions about model access, which we term Responsible Access Policies (RAPs). We recommend that, at a minimum, RAPs should include the following: i) processes for empirically evaluating model capabilities given different styles of access, ii) processes for assessing the risk profiles of different categories of user, and iii) clear and robust pre-commitments regarding when to grant or revoke specific types of access for particular groups under specified conditions.

cs.CY↗

arXiv2024

Governing dual-use technologies: Case studies of international security agreements and lessons for AI governance

International AI governance agreements and institutions may play an important role in reducing global security risks from advanced AI. To inform the design of such agreements and institutions, we conducted case studies of historical and contemporary international security agreements. We focused specifically on those arrangements around dual-use technologies, examining agreements in nuclear security, chemical weapons, biosecurity, and export controls. For each agreement, we examined four key areas: (a) purpose, (b) core powers, (c) governance structure, and (d) instances of non-compliance. From these case studies, we extracted lessons for the design of international AI agreements and governance institutions. We discuss the importance of robust verification methods, strategies for balancing power between nations, mechanisms for adapting to rapid technological change, approaches to managing trade-offs between transparency and security, incentives for participation, and effective enforcement mechanisms.

cs.CY↗

arXiv2024

Verification methods for international AI agreements

What techniques can be used to verify compliance with international agreements about advanced AI development? In this paper, we examine 10 verification methods that could detect two types of potential violations: unauthorized AI training (e.g., training runs above a certain FLOP threshold) and unauthorized data centers. We divide the verification methods into three categories: (a) national technical means (methods requiring minimal or no access from suspected non-compliant nations), (b) access-dependent methods (methods that require approval from the nation suspected of unauthorized activities), and (c) hardware-dependent methods (methods that require rules around advanced hardware). For each verification method, we provide a description, historical precedents, and possible evasion techniques. We conclude by offering recommendations for future work related to the verification and enforcement of international AI governance agreements.

cs.CY↗