Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Responsible AI”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 37 records · Page 2

AI Ethics Appendix: A Novel Approach to AI Ethics Workforce Development

The "AI Ethics Appendix" is a game-based design fiction for the deliberation of uncertain artificial intelligence (AI) ethical scenarios. The game is intended to be used as a tool for AI practitioners and industry professionals to grow responsible and ethical AI knowledge as they integrate this technology into their development. As AI/ML ethical considerations and governmental compliance develop, it is important to encourage teams to encourage teams to incorporate diverse thinking early in the development cycle and consider how different stakeholders may be affected by the technology. This game accomplishes these goals through storytelling and meaningful game interaction based on methods from game design as well as speculative and design fiction in the field of Human-Centered Design. The game mechanics were informed by the NASA Framework for the Ethical Use of Artificial Intelligence, Executive Order 13960, examples of AI use cases, and colleagues' work experiences with AI/ML.

trustworthy↗

The Crystal Structure of the Escherichia coli Autoinducer-2 Processing Protein LsrF

Many bacteria produce and respond to the quorum sensing signal autoinducer-2 (AI-2). Escherichia coli and Salmonella typhimurium are among the species with the lsr operon, an operon containing AI-2 transport and processing genes that are up regulated in response to AI-2. One of the Lsr proteins, LsrF, has been implicated in processing the phosphorylated form of AI-2. Here, we present the structure of LsrF, unliganded and in complex with two phospho-AI-2 analogues, ribose-5-phosphate and ribulose-5-phosphate. The crystal structure shows that LsrF is a decamer of (??)8-barrels that exhibit a previously unseen N-terminal domain swap and have high structural homology with aldolases that process phosphorylated sugars. Ligand binding sites and key catalytic residues are structurally conserved, strongly implicating LsrF as a class I aldolase.

36 MATERIALS SCIENCE↗

Digital Twin + AI: Control Room of the Future [Slides]

The control room functions as the central brain of the grid, essential for balancing supply and demand and ensuring moment-to-moment grid reliability. Like the human brain, which processes sensory data to make decisions, control room operators analyze operational data from power generation, transmission, and distribution to make informed decisions. Currently, decision-making primarily rests with operators due to hardware and software limitations. However, with technological advancements, Digital Twins and AI are becoming high interest points in the control room's decision-making pilot programs. NREL is developing a comprehensive decision-making platform that integrates Digital Twins, AI, and advanced visualization techniques. As this integration progresses, the role of Digital Twins will evolve from conducting automated simulations to serving as a Trustworthy AI enabler, offering verification and validation of AI-generated response for power systems or providing physics-aware synthetic data of AI pre-training.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Workflow Provenance in the Computing Continuum for Responsible, Trustworthy, and Energy-Efficient AI

As Artificial Intelligence (AI) becomes more pervasive in our society, it is crucial to develop, deploy, and assess Responsible and Trustworthy AI (RTAI) models, i.e., those that consider not only accuracy but also other aspects, such as explainability, fairness, and energy efficiency. Workflow provenance data have historically enabled critical capabilities towards RTAI. Provenance data derivation paths contribute to responsible workflows through transparency in tracking artifacts and resource consumption. Provenance data are well-known for their trustworthiness helping explainability, reproducibility, and accountability. However, there are complex challenges to achieve RTAI, which are further complicated by the heterogeneous infrastructure in the computing continuum (Edge-Cloud-HPC) used to develop and deploy models. As a result, a significant research and development gap remains between workflow provenance data management and RTAI. In this paper, we present a vision of the pivotal role of workflow provenance in supporting RTAI and discuss related challenges. We present a schematic view between RTAI and provenance, and highlight open research directions.

Santos Souza, Renan↗

Poster: Responsible Adoption of Artificial Intelligence (AI) in Electric Grid Operations

The rapid integration of artificial intelligence (AI) in the utility transmission and distribution (T&D) sector is revolutionizing traditional grid management practices. As utilities encounter complexities from evolving consumer behaviors and energy integration, AI becomes a critical solution for enhancing grid monitoring, fault detection, and operational optimization. However, increased reliance on interconnected technologies introduces significant cybersecurity risks, regulatory compliance challenges, and human factors concerns. This study proposes a strategic, responsible and consequence-driven approach to AI implementation, examining the dual nature of AI adoption by highlighting its transformative benefits for utilities and associated risks. It provides utilities with a framework for evaluating AI integration, enabling them to navigate challenges and capitalize on opportunities to achieve greater reliability, efficiency, and resilience in an increasingly complex energy landscape.

24 - POWER TRANSMISSION AND DISTRIBUTION↗

Agentic framework for programmatic crystal structure generation using a fine-tuned worker–supervisor large language model

Platinum group metals (PGMs) underpin many catalytic technologies but face severe supply constraints, motivating the search for alternative materials and computational methods to accelerate discovery. While atomistic simulation tools such as Pymatgen and ASE have streamlined structure manipulation, they require detailed inputs, limiting accessibility for experimentalists and slowing early-stage exploration. Here, in this study, we present an AI-driven agentic framework that orchestrates worker–supervisor large language models (LLMs). The worker translates natural-language prompts of varying abstraction into valid crystallographic structures using a compact LLM fine-tuned with low-rank adaptation on a curated text–code–CIF dataset, emphasizing energy-efficient training. Benchmarking against the baseline CodeGen-350M-mono model shows that fine-tuning reduces hallucination rates from 100% to as low as 5% and improves structural match accuracy to up to 82% for fully specified inputs. Accuracy declines with decreasing prompt detail but remains nontrivial even when only stoichiometry and space group are provided, underscoring the LLM’s capacity for crystallographic inference. The supervisor Claude LLM evaluates the outputs and triggers iterative refinement through the worker’s built-in structure manipulation capabilities (e.g., supercell scaling, strain, vacancy, and substitution operations). We further demonstrate use cases for technologically relevant catalysts, including IrO 2 , pyrochlore Pb 2 Ir 2 O 7 , Ni 2 FeO 4 , and Ni 3 Mo, where the framework generates physically consistent structures that can be refined via geometry optimization. This work introduces a low-energy, language-driven pathway for integrating human and machine intelligence in materials design, paving the way for AI-assisted synthesis planning and high-throughput screening of complex oxides.

AI agent↗

MLCommons Science Benchmarks

Benchmarks are a cornerstone of modern machine learning practice, providing standardized eval- uations that enable reproducibility, comparison, and scientific progress. Yet, as AI systems particularly deep learning models become increasingly dynamic, traditional static benchmarking approaches are losing their relevance. Models rapidly evolve in architecture, scale, and capability; datasets shift; and deployment contexts continuously change, creating a moving target for evaluation. Without adaptive benchmarking frame- works, both scientific assessment and real-world de- ployment risk becoming misaligned with actual system behavior. Drawing on our experience from MLCommons, educa- tional initiatives, and government programs such as the DOE s Million Parameter Consortium, we identify key barriers that hinder the broader adoption and utility of benchmarking in AI. These include substantial resource demands, limited access to specialized hardware, lack of expertise in benchmark design, and uncertainty among practitioners about how to relate benchmark results to their own application domains. Moreover, current benchmarks often emphasize peak performance on leadership-class hardware, offering limited guidance for more diverse, real-world deployment scenarios. We argue that benchmarking itself must become dy- namic in order to incorporate evolving models, updated data, and heterogeneous computational platforms while maintaining transparency, reproducibility, and inter- pretability. Democratizing this process requires not only technical innovation, but also systematic educational efforts spanning undergraduate to professional levels to develop sustained expertise in benchmark design and use. Finally, benchmarks should be framed and com- municated to support application-relevant comparisons, enabling both developers and users to make informed, context-sensitive decisions. Advancing dynamic and inclusive benchmarking practices will be essential to ensure that evaluation keeps pace with the evolving AI landscape and supports responsible, reproducible, and accessible AI deployment.

Hawks, Benjamin G. [Fermilab]↗

Application of a two-stream radiative transfer model for leaf lignin and cellulose concentrations from spectral reflectance measurements, part 1

Lignin and nitrogen contents of leaves constitute the primary rate-limiting parameters for the decomposition of forest litter, and are determinants of nutrient- and carbon-cyclic rates in forest ecosystems (Melillo et al., 1982). Wessman et al. (1988a) developed empirical multivariate relationships between forest canopy lignin amount and the (first-difference) AIS spectral response in three bands spread over the wavelength interval 1256-1555 nm. Wessman et al. (1988b) and McLellan et al. (1991) developed similar regression relationships from laboratory reflectance measurements on dried samples prepared in a standard fashion. They used four to six infrared bands for analysis of nitrogen, lignin and cellulose content of foliage in forest and prairie species. In the present article (Parts 1 and 2) the feasibility of compositional determinations is explored using positions of composite absorption bands that originate from mixtures of lignin, cellulose, and possibly other chemical constituents in the spectral reflectance of green leaves. To carry out this program, we employ full-spectral-resolution single-leaf diffuse reflectance measurements made with a laboratory spectrometer and integrating sphere. The leaf and other chemical reflectance data compiled by Elvidge (1990) have also been utilized extensively.

Conel, James E.↗

AI Benchmark Democratization and Carpentry

Benchmarks are a cornerstone of modern machine learning, enabling reproducibility, comparison, and scientific progress. However, AI benchmarks are increasingly complex, requiring dynamic, AI-focused workflows. Rapid evolution in model architectures, scale, datasets, and deployment contexts makes evaluation a moving target. Large language models often memorize static benchmarks, causing a gap between benchmark results and real-world performance. Beyond traditional static benchmarks, continuous adaptive benchmarking frameworks are needed to align scientific assessment with deployment risks. This calls for skills and education in AI Benchmark Carpentry. From our experience with MLCommons, educational initiatives, and programs like the DOE's Trillion Parameter Consortium, key barriers include high resource demands, limited access to specialized hardware, lack of benchmark design expertise, and uncertainty in relating results to application domains. Current benchmarks often emphasize peak performance on top-tier hardware, offering limited guidance for diverse, real-world scenarios. Benchmarking must become dynamic, incorporating evolving models, updated data, and heterogeneous platforms while maintaining transparency, reproducibility, and interpretability. Democratization requires both technical innovation and systematic education across levels, building sustained expertise in benchmark design and use. Benchmarks should support application-relevant comparisons, enabling informed, context-sensitive decisions. Dynamic, inclusive benchmarking will ensure evaluation keeps pace with AI evolution and supports responsible, reproducible, and accessible AI deployment. Community efforts can provide a foundation for AI Benchmark Carpentry.

von Laszewski, Gregor [Virginia U.]↗

Involvement of Shewanella oneidensis MR-1 LuxS in Biofilm Development and Sulfur Metabolism

The role of LuxS in Shewanella oneidensis MR-1 has been examined by transcriptomic profiling, biochemical, and physiological experiments. The results indicate that a mutation in luxS alters biofilm development, not by altering quorum-sensing abilities but by disrupting the activated methyl cycle (AMC). The S. oneidensis wild type can produce a luminescence response in the AI-2 reporter strain Vibrio harveyi MM32. This luminescence response is abolished upon the deletion of luxS. The deletion of luxS also alters biofilm formations in static and flowthrough conditions. Genetic complementation restores the mutant biofilm defect, but the addition of synthetic AI-2 has no effect. These results suggest that AI-2 is not used as a quorum-sensing signal to regulate biofilm development in S. oneidensis. Growth on various sulfur sources was examined because of the involvement of LuxS in the AMC. A mutation in luxS produced a reduced ability to grow with methionine as the sole sulfur source. Methionine is a key metabolite used in the AMC to produce a methyl source in the cell and to recycle homocysteine. These data suggest that LuxS is important to metabolizing methionine and the AMC in S. oneidensis.

59 BASIC BIOLOGICAL SCIENCES↗

ADEPT: A Pedagogical Framework for Integrating Agentic AI with Deterministic Scientific Workflows

The integration of Large Language Models (LLMs) into scientific research promises to accelerate discovery, yet a significant gap remains between the dynamic reasoning of Artificial Intelligence (AI) agents and the static, deterministic nature of canonical scientific workflows. This paper introduces ADEPT (Agentic Discovery and Exploration Platform for Tools), a reference architecture and pedagogical framework explicitly designed to bridge this gap. ADEPT's primary mission is to provide a transparent, "glass-box" environment where researchers and engineers can learn to effectively wrap established scientific software (e.g., BLAST, Nextflow pipelines) and compose it into reliable, agent-driven workflows. We describe its modular, multi-server architecture, which leverages the Model Context Protocol (MCP) for tool serving, LangGraph for robust agentic orchestration, and a secure nsjail-based sandbox for safe code execution. By prioritizing architectural clarity, safety, and modularity, ADEPT serves as an extensible blueprint for building trustworthy AI-augmented systems and fosters the collaborative development necessary to responsibly employ agentic AI for science. We provide practical examples of how to adapt and extend this framework, highlighting its utility in workforce development and AI-readiness capabilities across research and development projects.

97 MATHEMATICS AND COMPUTING↗

A Fire Community Observatory: Interdisciplinary, AI-informed Post-Fire Rapid Response for Improved Water Cycle Science at Watershed Scale

Wildfire is an ecological disturbance that disrupts the hydrological cycle. In the past few years, a record number of multiple-and-compounding fires have occurred across urban-wildland gradients in the Western United States. Changes to watershed hydrological partitioning in response to fires (infiltration, runoff, evapotranspiration) presents unprecedented challenges to “Water-in-the-West” through negative impacts to water supply and its quality, and is a direct threat to downstream communities, groundwater, and drinking water supply infrastructure. While much work is being done to advance Artificial Intelligence and Machine Learning (AI/ML) use during fires for emergency response (i.e. predict fire movement, direct evacuations), significant potential exists to use AI/ML to address three scientific grand challenges that are rarely addressed in a convergent science context: 1) how to enhance the potential resiliency of a landscape before fire(s), 2) how to cost-effectively and optimally monitor watershed changes after fires, and 3) how to predict future hydrological and biogeochemical trajectories in fire-impacted watershed given climate change. This whitepaper addresses DOE Focal Area 1: Data acquisition and assimilation enabled by machine learning, AI, and advanced methods.

58 GEOSCIENCES↗

Observational Capabilities to Capture Water Cycle Event Dynamics and Impacts in the Age of AI

This whitepaper is responsive to focal area (1) Data acquisition and assimilation enabled by machine learning (ML), Artificial Intelligence (AI), and advanced methods. Here we describe how Earth observations specific to water cycle disturbances can be collected in parallel with and integrated into future model development, and make use of the latest technologies other than AI/ ML such as 5G/satellite, edge computing, big data technologies, and cloud computing.

58 GEOSCIENCES↗

AI@DOE Interim Executive Report

Interim executive report for DOE’s Office of Science, NNSA, and Applied Energy Offices, in collaboration with the Artificial Intelligence and Technology Office, DOE AI roundtable workshops (December 2021 through February 2022) (“AI@DOE”). These workshops were used to help identify AI research requirements and priorities for the next decade for Department of Energy in developing machine-learning-based prediction and decision-support capabilities that address and anticipate DOE mission challenges. These missions’ challenges for AI include: AI to advance scientific discovery and technological innovation AI to support high consequence decisions while managing risk AI to improve DOE’s responsiveness to national & global issues AI to assist stewardship of the environment and national critical infrastructure These will be addressed in large part through the development and application of new and powerful artificial intelligence and/or machine learning algorithms and strategies. https://web.cvent.com/event/fc3922f8-fc75-4041-a317-f13a1da44f7c/summary?locale=en-US&i=2Wx2tjbStkydRFq_N2whPw

97 MATHEMATICS AND COMPUTING↗

Compound gravity receptor polarization vectors evidenced by linear vestibular evoked potentials

The utricle and saccule are gravity receptor organs of the vestibular system. These receptors rely on a high-density otoconial membrane to detect linear acceleration and the position of the cranium relative to Earth's gravitational vector. The linear vestibular evoked potential (VsEP) has been shown to be an effective non-invasive functional test specifically for otoconial gravity receptors (Jones et al., 1999). Moreover, there is some evidence that the VsEP can be used to independently test utricular and saccular function (Taylor et al., 1997; Jones et al., 1998). Here we characterize compound macular polarization vectors for the utricle and saccule in hatchling chickens. Pulsed linear acceleration stimuli were presented in two axes, the dorsoventral (DV, +/- Z axis) to isolate the saccule, and the interaural (IA, +/- Y axis) to isolate the utricle. Traditional signal averaging was used to resolve responses recorded from the surface of the skull. Latency and amplitude of eighth nerve components of the linear VsEP were measured. Gravity receptor responses exhibited clear preferences for one stimulus direction in each axis. With respect to each utricular macula, lateral translation in the IA axis produced maximum ipsilateral response amplitudes with substantially greater amplitude intensity (AI) slopes than medially directed movement. Downward caudal motions in the DV axis produced substantially larger response amplitudes and AI slopes. The results show that the macula lagena does not contribute to the VsEP compound polarization vectors of the sacculus and utricle. The findings suggest further that preferred compound vectors for the utricle depend on the pars externa (i.e. lateral hair cell field) whereas for the saccule they depend on pars interna (i.e. superior hair cell fields). These data provide evidence that maculae saccule and utricle can be selectively evaluated using the linear VsEP.

NASA Discipline Developmental Biology↗

Workshop Summary Report on Using AI Tools to Improve the Efficiency and Outcomes of the NEPA Process: AI for Permitting Workshop at the 2025 National Association of Environmental Professionals (NAEP) Annual Conference

On April 29, 2025, the U.S. Department of Energy and Pacific Northwest National Laboratory hosted a workshop at the National Association of Environmental Professionals 2025 Conference and Training Symposium in Charleston, South Carolina, titled, “Effective and Responsible Use of Customized AI Tools to Improve the Efficiency and Outcomes of the NEPA Process.” The objectives of this workshop were to make environmental practitioners aware of the potential for using artificial intelligence in the National Environmental Policy Act process, demonstrate examples of how artificial intelligence can be integrated effectively to improve efficiency and outcomes and solicit questions and feedback from practitioners. This report summarizes the key points from all talks and case studies, as well as audience questions and feedback on the presentation topics and the broader topic of "AI in permitting". The report concludes by highlighting the key barriers and opportunities for the implementation of AI in permitting, as discussed during the workshop.

54 ENVIRONMENTAL SCIENCES↗

From Rules to Reasoning: A Survey of Large Language Model-Based Approaches to Scientific Hypothesis and Idea Generation

Scientific hypothesis generation represents a fundamental challenge in contemporary research due to exponentially expanding literature volumes and increasing disciplinary specialization. Large language models (LLMs) have emerged as transformative tools for automated scientific discovery, moving beyond traditional rule-based and literature-mining approaches. Four paradigmatic approaches define current LLM-driven hypothesis generation: direct prompting and fine-tuning methods, knowledge-enhanced frameworks integrating retrieval-augmented generation (RAG), multi-agent collaborative systems simulating research teams, and reasoning-focused approaches implementing cognitive architectures. Domain-specific applications demonstrate statistical equivalence to human expert performance in social psychology, experimental validation in biomedical research, and near-expert quality in astronomy. Evaluation methodologies encompass human expert assessment, LLM-as-judge frameworks, and comprehensive benchmarking systems. Technical challenges include hallucination management, knowledge integration limitations, and balancing novelty with feasibility. Future directions emphasize hybrid neural-symbolic architectures and sophisticated human-AI collaboration models for responsible scientific discovery acceleration.

AI-driven discovery↗