Pith. sign in

REVIEW 3 major objections 4 minor 186 references

This survey argues that governing Physical AI means treating governance as a life-cycle-spanning concern rather than a post-deployment compliance step, and offers a unified five-principle, five-stage framework to make that operational.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

T0 review · deepseek-v4-flash

2026-08-01 04:15 UTC pith:JI7QXNO2

load-bearing objection Useful survey of Physical AI governance, but the unified P-Gov/E-PAL framework is not yet demonstrated—the 15 vs 17 sub-component count and the ad hoc operationalization mapping need fixing before this becomes the reference it aims to be. the 3 major comments →

arxiv 2607.22877 v1 pith:JI7QXNO2 submitted 2026-07-24 cs.AI cs.HCcs.RO

Physical AI Governance: From Theory to Practice Across Life Cycle

Wang Yang , Shaobo Wang , Hongxuan Liu , Xiaoran Cai , Yunyu He , Jingzong Zhou , Mengzhong Ma , Yi Yu
show 3 more authors
Rohit Sharma Jingjing Fu Peng Qi
This is my paper
classification cs.AI cs.HCcs.RO
keywords Physical AIAI governanceembodied AIgovernance lifecyclesafetytrustworthy AIroboticsaccountability
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper is trying to establish that governance of Physical AI—embodied systems that perceive and act in the physical world—cannot be a checklist appended after deployment. It argues that existing AI governance frameworks, built for screen-based digital AI, miss what embodiment changes: real-time safety constraints, continuous interaction with dynamic environments, and the immediacy of physical harm. To close that gap, it proposes a unified framework, P-Gov, organizing the field into five principles (robust and safe operation, human-centered values, integrity/privacy/equity, accountability and oversight, sustainability) with sub-components, and a five-stage lifecycle, E-PAL (research, design, data, model, deployment), showing how each principle takes a stage-specific form at each step. If right, the contribution is a shared vocabulary and an operational checklist for a fragmented governance landscape—useful to researchers, developers, and policymakers, and a foundation for later, more quantitative governance metrics.

Core claim

The central claim is that 'governing Physical AI requires treating governance not as a compliance step appended after deployment, but as a concern that spans the entire life cycle of an embodied system.' The paper supports this by deriving two artifacts: a governance framework (P-Gov) with five principles and a set of sub-components, and an end-to-end lifecycle (E-PAL) with five stages. The framework is explicitly descriptive rather than causal: satisfying every element does not guarantee a well-governed system, but omitting the stage-specific form of an element leaves a corresponding governance gap. The survey then operationalizes each principle at each stage through concrete practices—sim-

What carries the argument

The machinery is a two-part taxonomy. P-Gov (Physical AI Governance Framework) is the first part: five principles—Robust & Safe Operation, Human-Centered Values, Integrity/Privacy/Equity, Accountability & Oversight, Sustainability—each decomposed into sub-components (the paper says fifteen; its own figure shows seventeen), each anchored to existing standards and literature. E-PAL (End-to-End Physical AI Lifecycle) is the second part: research, design, data, model, deployment, arranged into a knowledge-generation layer and a build-and-operation layer with feedback loops. The work these artifacts do is to give every governance concern a home: a principle tells you what to care about, a lifecyc

Load-bearing premise

The framework's categories are assumed to be a stable, complete, non-redundant synthesis of the governance literature—the paper does not describe a survey methodology, and its own text lists fifteen sub-components while its figure shows seventeen.

What would settle it

Compare the taxonomy against a systematically collected corpus of Physical AI governance documents (standards, incident reports, regulatory guidance). If a substantial set of governance concerns cannot be mapped to any P-Gov sub-component, or if two sub-components consistently map to the same concern, the claim that this is a unified synthesis fails. A simpler check: resolve the paper's own fifteen-versus-seventeen discrepancy; if the sub-component list is unstable within a single document, the framework is not yet a reliable synthesis.

Watch this falsifier — get emailed when new claim-graph text bears on it.

If this is right

  • Adopting the framework means safety, fairness, and accountability are no longer one-size-fits-all: each lifecycle stage carries its own concrete instantiation, from safety-by-design at the design stage to runtime assurance and resilience monitoring at deployment.
  • The three-tier structure gives regulators and developers a common map for locating governance gaps—if a stage-specific form of a principle is missing, there is an identifiable hole in the system's governance coverage.
  • The lifecycle framing makes governance inherently iterative: feedback from deployment operation flows back into research and design, so governance itself becomes a closed-loop process rather than a one-time certification.
  • The survey's catalog of concrete practices (sim-to-real reliability testing, data provenance, explainable and bias-audited policies, resilience monitoring, life-cycle sustainability assessment) can serve as a starting checklist for engineering teams building embodied systems.
  • The framework's descriptive, gap-based logic implies that a system can be well-governed only if every principle is explicitly addressed at every relevant stage; the paper frames omissions as governance gaps, not merely as best-effort shortfalls.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The framework's own instability—fifteen sub-components stated in the text but seventeen shown in the figure—suggests the taxonomy is still being actively reshaped; a formal ontology or inter-rater reliability study of the P-Gov categories would be a natural next step to test whether the categories are genuinely non-redundant and complete.
  • Because the paper stops short of metrics, its own future-work direction could be turned into a falsifiable test: define stage-specific quantitative indicators for each principle and compare development efforts that use the E-PAL checklist against those using generic AI governance lists, measuring post-incident failure rates.
  • The lifecycle framing could double as a risk-register template for insurers and regulators, converting each stage-specific governance element into an auditable control—an extension the paper does not explicitly draw but that follows directly from its operationalization drive.
  • The paper's emphasis on 'omitting the stage-specific form of an element leaves a governance gap' invites a concrete audit procedure: for a given embodied system, walk the E-PAL stages and check whether each of the five principles has a defined practice at each stage; any missing pair is a candidate gap.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit.

Referee Report

3 major / 4 minor

Summary. This paper presents a survey of governance for Physical AI—embodied AI systems that act in the physical world—and proposes two linked frameworks. First, P-Gov (called PAL-GF in Figure 2) organizes governance principles into five categories and, according to the text, fifteen sub-components (though Figure 2 lists seventeen). Second, E-PAL defines a five-stage lifecycle (Research, Design, Data, Model, Deployment) and maps governance practices onto each stage, including a three-tier structure (Fundamental, Knowledge Generation, Building/Deployment/Operation). The paper claims that these frameworks together provide a systematic foundation for making Physical AI safe, trustworthy, and socially beneficial by connecting high-level principles to concrete engineering practices. The conclusion explicitly hedges that the framework is descriptive rather than causal. The central contribution is the unified vocabulary and operational checklist it offers to a fragmented governance literature.

Significance. If the proposed frameworks are internally coherent and genuinely synthesized from the literature, this survey would be a useful reference for researchers, developers, and policymakers. The paper has notable strengths: it is carefully hedged about the limits of its claims; it grounds each governance element in cited studies and standards; it connects principles to concrete practices (e.g., sim-to-real reliability testing, data provenance, runtime resilience monitoring); and it explicitly frames the lifecycle as iterative. The inclusion of both a principle taxonomy and a lifecycle operationalization is timely and addresses a real gap. However, the value of the contribution rests almost entirely on the consistency and traceability of the framework itself. The paper's internal inconsistencies—particularly the fifteen-versus-seventeen sub-component discrepancy and the lack of an explicit mapping between P-Gov principles and the lifecycle governance elements—currently prevent the central claim from being established. These issues are fixable, but they are load-bearing rather than cosmetic.

major comments (3)
  1. [§2.1 vs. Figure 2] Section 2.1 states that P-Gov covers 'five fundamental governance principles and fifteen governance sub-components,' while Figure 2's caption says 'Seventeen governance sub-components are grouped into five categories.' This is not a minor count typo: the taxonomy itself is the paper's central contribution, and the mismatch indicates the set of sub-components is not fixed. Please reconcile the count and state the exact list. If some sub-components in Figure 2 are meant to be grouped differently, the grouping rule should be stated explicitly.
  2. [§4.2.1–4.2.6 and Figure 4] The operationalization layer is not systematically derived from the P-Gov taxonomy. Figure 2 includes seventeen sub-components (e.g., Adaptability, Accessibility, Control, Auditability, Resource Efficiency, Circularity), but §4.2 adds elements not present in that list: Reliability and Reproducibility (§4.2.2), Inclusiveness and Generalizability (§4.2.3), Traceability (§4.2.4), and Resilience (§4.2.6). Conversely, several P-Gov sub-components—Accessibility, Control, Auditability, Resource Efficiency, Circularity, among others—do not appear as explicit entries in Figure 4 or in any §4.2 subsection. The paper implies these are merely different aggregations (Section 4.2.1 calls some elements 'Foundational Governance Factors'), but no mapping rule is provided. Without a table or explicit derivation linking each P-Gov sub-component to the lifecycle-specific governance elements, the claim that
  3. [§4.1 and §2] The paper describes itself as a 'comprehensive survey' and claims to 'distill' the lifecycle and 'synthesize' governance principles, but it describes no survey methodology—no search strategy, inclusion/exclusion criteria, or procedure for selecting the five lifecycle stages and the governance categories. This is load-bearing because the validity of the proposed framework depends on the representativeness of the synthesis. If the categories are an arbitrary selection, the framework is a restatement of existing lists rather than a systematic foundation. At minimum, state how the five stages and the five principles were derived from the literature, or explicitly reposition the contribution as a proposal rather than a survey.
minor comments (4)
  1. [§4.2.1] The 'Human Value Alignment' subsection contains two consecutive paragraphs beginning 'For deployment,' which appears to be a structural duplication. The second 'For deployment' paragraph should be integrated or given a different subheading.
  2. [§4.2.5] Typo: 'eplainable' should be 'explainable' in the first sentence of the Model Governance Factors subsection.
  3. [§4.2.1] Typo: 'SAFETextare develeped' should be 'SAFEText are developed.'
  4. [Figure 4] Figure 4's caption could be clearer about whether the listed items are additional governance elements or instances of the P-Gov sub-components. A legend or note about the mapping to Figure 2 would help.

Circularity Check

0 steps flagged

No significant circularity: the survey's framework is descriptive, externally anchored, and lacks any derivation or prediction that reduces to its inputs.

full rationale

The paper is a literature survey, not a derivation. It contains no equations, fitted parameters, or quantitative predictions whose value could be forced by construction. Its central contribution—the P-Gov principle taxonomy and the E-PAL lifecycle—is presented explicitly as a synthesis of external standards and literature (ISO, NIST, IEEE, and numerous third-party cited works), not as a consequence of the authors' own prior results. The self-citations that appear (Li et al. 2023; Cai et al. 2026) are used only as background or comparative references for traditional AI and do not carry the framework's argument. The observed internal inconsistencies, such as the 15-versus-17 sub-component count and the appearance of stage-specific governance elements not present in Figure 2, are taxonomy and completeness concerns rather than circularity: no step in the paper claims to predict or derive one from the other by construction. The conclusion further disclaims causal force, stating that the framework is 'descriptive rather than causal' and only identifies omitted stage-specific forms as governance gaps. No circular step can be exhibited with a specific reduction, so the honest finding is no significant circularity.

Axiom & Free-Parameter Ledger

0 free parameters · 4 axioms · 0 invented entities

The paper's central claim rests on the assumption that a small set of governance principles and a five-stage lifecycle can organize the whole field of Physical AI governance. These are domain-level framing assumptions rather than fitted parameters or new physical entities. No free parameters are involved because no data are fit. The only 'invented' items are the named frameworks P-Gov and E-PAL, which are conceptual organizations, not postulates requiring independent empirical evidence.

axioms (4)
  • domain assumption Physical AI is a distinct category of AI, distinct from both digital AI and robotics, warranting a dedicated governance framework.
    The entire paper is predicated on the claim that Physical AI (embodied systems) introduces governance challenges not covered by traditional AI governance (Sec. 1, Intro). If this category distinction is not meaningful, the survey's motivation collapses.
  • domain assumption The surveyed literature is representative and sufficient to support the taxonomy.
    The paper draws its principles and lifecycle stages from about 150 cited works, but it does not describe a systematic search or selection methodology (Sec. 2). The taxonomy is only as good as the chosen literature set.
  • domain assumption Existing AI governance practices can be extended to Physical AI with augmentation.
    In Sec. 4.2, the paper states that 'Many existing AI governance practices can be extended to Physical AI; however, they must be augmented.' This assumption underlies every operationalization recommendation.
  • domain assumption The five-stage lifecycle (research, design, data, model, deployment) is a valid universal representation of Physical AI systems.
    The paper distills the lifecycle into these five stages (Sec. 4.1) and then organizes all governance analysis around them. If a significant portion of Physical AI development does not fit this lifecycle, the operationalization mapping is incomplete.

pith-pipeline@v1.3.0-alltime-deepseek · 33507 in / 8425 out tokens · 75930 ms · 2026-08-01T04:15:00.486530+00:00 · methodology

0 comments
read the original abstract

With the emergence of Physical AI, artificial intelligence is extending beyond screen-based applications to embodied systems that perceive, interact with, and act in the physical world. Unlike traditional AI, Physical AI operates under real-time safety constraints, continuously interacts with dynamic environments, and coexists with humans, introducing governance challenges that existing AI governance frameworks do not explicitly address. This paper presents a comprehensive survey of Physical AI governance from both scientific and operational perspectives. We synthesize existing governance principles and organize them into a unified governance framework tailored to physical AI systems. Building on this foundation, we propose a five-stage Physical AI lifecycle comprising research, design, data, model development, and deployment, and demonstrate how governance can be operationalized across each stage through concrete implementation practices. By connecting governance principles with engineering workflows, this survey provides a structured reference for researchers, developers, and policymakers to build Physical AI systems that are safe, trustworthy, and aligned with societal values.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

186 extracted references · 8 canonical work pages · 2 internal anchors

  1. [1]

    2026 , month =

    VerWey, John , title =. 2026 , month =

    2026

  2. [2]

    Journal of Computer Science and Technology , year =

    En-Hua Wu and You-Quan Liu and Tian-Chen Xu and Li-Xin Ren and Yi-Ming Qin and Ming-Yu Wei and Xiao-Wei He and Dong-Yan Yuan and Wen-Chao Hou and Zhi-Wei Ma and Bin Sheng , title =. Journal of Computer Science and Technology , year =. doi:10.1007/s11390-026-6258-x , publisher =

    work page doi:10.1007/s11390-026-6258-x

  3. [3]

    1948 , publisher=

    Cybernetics: Or Control and Communication in the Animal and the Machine , author=. 1948 , publisher=

    1948

  4. [4]

    , title =

    Nilsson, Nils J. , title =

  5. [5]

    Extreme Mechanics Letters , volume=

    Physical Intelligence as a New Paradigm , author=. Extreme Mechanics Letters , volume=. 2021 , publisher=

    2021

  6. [6]

    2025 , month = dec, note =

    Physical AI and Humanoid Robots , howpublished =. 2025 , month = dec, note =

    2025

  7. [7]

    Software Engineering and Formal Methods (SEFM 2025) , pages =

    Yelyzaveta Kurkchi and Catia Trubiani , title =. Software Engineering and Formal Methods (SEFM 2025) , pages =. 2025 , publisher =

    2025

  8. [8]

    Intelligent Systems with Applications , volume =

    Pranjal Biswas and Thomas Minhyung Kim and Woo Soo Kim , title =. Intelligent Systems with Applications , volume =. 2026 , publisher =

    2026

  9. [9]

    arXiv preprint arXiv:2605.02900 , year =

    Safety in Embodied AI: A Survey of Risks, Attacks, and Defenses , author =. arXiv preprint arXiv:2605.02900 , year =

    Pith/arXiv arXiv

  10. [10]

    2026 , eprint =

    World Model for Robot Learning: A Comprehensive Survey , author =. 2026 , eprint =

    2026

  11. [11]

    IEEE Transactions on Pattern Analysis and Machine Intelligence , year =

    Aligning Perception, Reasoning, Modeling and Interaction: A Survey on Physical AI , author =. IEEE Transactions on Pattern Analysis and Machine Intelligence , year =

  12. [12]

    2025 , url =

    AI Governance Index 2025: A Report into the State of AI Governance in Enterprise , institution =. 2025 , url =

    2025

  13. [13]

    Advanced Robotics , volume =

    Hentout, Abdelfetah and Aouache, Mustapha and Maoudj, Abderraouf and Akli, Isma , title =. Advanced Robotics , volume =. 2019 , doi =

    2019

  14. [14]

    IEEE Transactions on Software Engineering , volume =

    Jiayang Song and Xuan Xie and Lei Ma , title =. IEEE Transactions on Software Engineering , volume =. 2023 , doi =

    2023

  15. [15]

    2022 , institution =

    2022

  16. [16]

    2019 , institution =

    2019

  17. [17]

    2023 , institution =

    2023

  18. [18]

    Vasilakos and Qi Hao and Rui Ma and Yogendra Patil and Ting Zhang and Jiang Lu and Xin Li and Neal N

    Fei Hu and Yu Lu and Athanasios V. Vasilakos and Qi Hao and Rui Ma and Yogendra Patil and Ting Zhang and Jiang Lu and Xin Li and Neal N. Xiong , title =. Future Generation Computer Systems , volume =. 2016 , month = mar, doi =

    2016

  19. [19]

    Annaswamy and Karl Henrik Johansson and Aranya Chakrabortty , title =

    Seyed Mehran Dibaji and Mohammad Pirani and David Bezalel Flamholz and Anuradha M. Annaswamy and Karl Henrik Johansson and Aranya Chakrabortty , title =. Annual Reviews in Control , volume =. 2019 , doi =

    2019

  20. [20]

    ACM Transactions on Cyber-Physical Systems , volume =

    Sanghoon Lee and Jiyeong Chae and Haewon Jeon and Taehyun Kim and Yeong-Gi Hong and Doo-Sik Um and Taewoo Kim and Kyung-Joon Park , title =. ACM Transactions on Cyber-Physical Systems , volume =. 2025 , month = apr, publisher =

    2025

  21. [21]

    ACM Transactions on Software Engineering and Methodology , volume =

    Hugo Araujo and Mohammad Reza Mousavi and Mahsa Varshosaz , title =. ACM Transactions on Software Engineering and Methodology , volume =. 2023 , publisher =. doi:10.1145/3542945 , url =

    work page doi:10.1145/3542945 2023

  22. [22]

    Frontiers in Robotics and AI , volume =

    Yunwei Zhang and Jing Tian and Qiaochu Xiong , title =. Frontiers in Robotics and AI , volume =. 2025 , doi =

    2025

  23. [23]

    IEEE Internet of Things Journal , volume =

    Lei Tang and Ruijie Wang and Zhanwen Liu and Yunji Liang and Yuanyuan Niu and Wei Zhu and Wenhui Zhu and Zongtao Duan , title =. IEEE Internet of Things Journal , volume =. 2025 , month = jan, doi =

    2025

  24. [24]

    ACM Transactions on Human-Robot Interaction , volume =

    Saurav Singh and Esa Rantanen and Jamison Heard , title =. ACM Transactions on Human-Robot Interaction , volume =. 2026 , publisher =. doi:10.1145/3776548 , url =

    doi:10.1145/3776548 2026

  25. [25]

    Frontiers in Robotics and AI , volume =

    Wenzheng Zhao and Kruthika Gangaraju and Fengpei Yuan , title =. Frontiers in Robotics and AI , volume =. 2025 , month = aug, doi =

    2025

  26. [26]

    ACM Computing Surveys , volume =

    Mariana Segovia-Ferreira and Jose Rubio-Hernan and Ana Rosa Cavalli and Joaquin Garcia-Alfaro , title =. ACM Computing Surveys , volume =. 2024 , doi =

    2024

  27. [27]

    AIS Transactions on Human-Computer Interaction , volume =

    Ben Shneiderman , title =. AIS Transactions on Human-Computer Interaction , volume =. 2020 , doi =

    2020

  28. [28]

    Annual Reviews in Control , volume=

    From ethics to standards -- A path via responsible AI to cyber-physical production systems , author=. Annual Reviews in Control , volume=. 2022 , publisher=

    2022

  29. [29]

    Future Internet , volume=

    Security and Privacy in Physical--Digital Environments: Trends and Opportunities , author=. Future Internet , volume=. 2025 , publisher=

    2025

  30. [30]

    Internet of Things , volume=

    A Unified Privacy Preserving Model with AI at the Edge for Human-in-the-Loop Cyber-Physical Systems , author=. Internet of Things , volume=. 2024 , publisher=

    2024

  31. [31]

    arXiv preprint arXiv:2605.05017 , year=

    Position: Embodied AI Requires a Privacy-Utility Trade-off , author=. arXiv preprint arXiv:2605.05017 , year=

    Pith/arXiv arXiv

  32. [32]

    Preprints , year=

    A Taxonomy of AI Techniques for Security and Privacy in Cyber-Physical Systems , author=. Preprints , year=

  33. [33]

    Proceedings of the 2022 ACM Conference on Fairness, Accountability, and Transparency (FAccT '22) , year =

    Hundt, Andrew and Agnew, William and Zeng, Vicky and Kacianka, Severin and Gombolay, Matthew , title =. Proceedings of the 2022 ACM Conference on Fairness, Accountability, and Transparency (FAccT '22) , year =. doi:10.1145/3531146.3533138 , isbn =

    arXiv 2022

  34. [34]

    Ramos and Keerthi Sagar and Aswin Ramasubramanian and John D

    Shakra Mehak and Inês F. Ramos and Keerthi Sagar and Aswin Ramasubramanian and John D. Kelleher and Michael Guilfoyle and Gabriele Gianini and Ernesto Damiani and Maria Chiara Leva , title =. Frontiers in Robotics and AI , volume =. 2024 , doi =

    2024

  35. [35]

    Automation , year =

    Maral Niazi and Hossein Hassani and Madison Lee , title =. Automation , year =

  36. [36]

    Chemweno and Kostas Nizamis and Eduard Fosch-Villaronga , title =

    Alberto Martinetti and Peter K. Chemweno and Kostas Nizamis and Eduard Fosch-Villaronga , title =. Frontiers in Chemical Engineering , volume =. 2021 , doi =

    2021

  37. [37]

    Deep Science Publishing , volume =

    Birupaksha Biswas and Suhena Sarkar , title =. Deep Science Publishing , volume =. 2026 , doi =

    2026

  38. [38]

    Proceedings of the IEEE International Conference on Robotics and Automation (ICRA) , pages =

    Shengduo Chen and Yaowei Sun and Dachuan Li and Qiang Wang and Qi Hao and Joseph Sifakis , title =. Proceedings of the IEEE International Conference on Robotics and Automation (ICRA) , pages =. 2021 , publisher =

    2021

  39. [39]

    Kochenderfer and Marco Pavone , title =

    Rachel Luo and Shengjia Zhao and Yixuan Wang and Yifei Wang and Chelsea Finn and Mykel J. Kochenderfer and Marco Pavone , title =. The International Journal of Robotics Research , volume =. 2024 , doi =

    2024

  40. [40]

    Communications of the ACM , year =

    Yung-Chen Tang and Pin-Yu Chen and Tsung-Yi Ho , title =. Communications of the ACM , year =

  41. [41]

    Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing (EMNLP) , pages =

    Sharon Levy and Emily Allaway and Melanie Subbiah and Lydia Chilton and Desmond Patton and Kathleen McKeown and William Yang Wang , title =. Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing (EMNLP) , pages =. 2022 , publisher =

    2022

  42. [42]

    IEEE Access , volume =

    Yuchen Jiang and Shen Yin and Okyay Kaynak , title =. IEEE Access , volume =. 2018 , doi =

    2018

  43. [43]

    Entropy , volume =

    Oleg Illiashenko and Vyacheslav Kharchenko and Ievgen Babeshko and Herman Fesenko and Felicita Di Giandomenico , title =. Entropy , volume =. 2023 , doi =

    2023

  44. [44]

    Proceedings of the 13th ACM/IEEE International Conference on Cyber-Physical Systems (ICCPS) , pages =

    Xiangguo Liu and Chao Huang and Yixuan Wang and Bowen Zheng and Qi Zhu , title =. Proceedings of the 13th ACM/IEEE International Conference on Cyber-Physical Systems (ICCPS) , pages =. 2022 , publisher =

    2022

  45. [45]

    Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence (IJCAI) , pages =

    Takeshi Kojima and Yaonan Zhu and Yusuke Iwasawa and Toshinori Kitamura and Gang Yan and Shu Morikuni and Ryosuke Takanami and Alfredo Solano and Tatsuya Matsushima and Akiko Murakami and Yutaka Matsuo , title =. Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence (IJCAI) , pages =. 2025 , doi =

    2025

  46. [46]

    and others , title =

    Zhang, Y. and others , title =. arXiv preprint arXiv:2505.15471 , year =

    Pith/arXiv arXiv

  47. [47]

    Human-Collaborative Artificial Intelligence Along With Social Values in Industry 5.0: A Survey of the State-of-the-Art , journal =

    Mahdi Khosravy and Neeraj Gupta and Antoine Pasquali and Nilanjan Dey and Rub. Human-Collaborative Artificial Intelligence Along With Social Values in Industry 5.0: A Survey of the State-of-the-Art , journal =. 2024 , month = feb, publisher =

    2024

  48. [48]

    AI Perspectives , volume =

    Stephane Doncieux and Raja Chatila and Sirko Straube and Frank Kirchner , title =. AI Perspectives , volume =. 2022 , month = jan, doi =

    2022

  49. [49]

    MIT Sloan Management Review , year =

    Benedict Jun Ma and Maria Jesus Saenz , title =. MIT Sloan Management Review , year =

  50. [50]

    Winfield, Alan F. T. and Jirotka, Marina , title =. Philosophical Transactions of the Royal Society A , volume =

  51. [51]

    International Journal of Social Robotics , year =

    Giulio Antonio Abbo and Tony Belpaeme and Micol Spitale , title =. International Journal of Social Robotics , year =

  52. [52]

    Johnson and Luis Garcia and Taylor T

    Stanley Bak and Taylor T. Johnson and Luis Garcia and Taylor T. Johnson and others , title =. ACM Transactions on Cyber-Physical Systems , year =

  53. [53]

    Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops , pages =

    Vasu Singh and Siva Kumar Sastry Hari and Timothy Tsai and Mandar Pitale , title =. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops , pages =. 2021 , publisher =

    2021

  54. [54]

    Neu , title =

    Fariborz Farahmand and Richard W. Neu , title =. Fatigue & Fracture of Engineering Materials & Structures , year =. doi:10.1111/ffe.14575 , note =

    work page doi:10.1111/ffe.14575

  55. [55]

    Ramírez-Duque and Mary Ellen Foster , title =

    Nicholas Callander and Andrés A. Ramírez-Duque and Mary Ellen Foster , title =. Companion of the 2024 ACM/IEEE International Conference on Human-Robot Interaction (HRI '24 Companion) , pages =. 2024 , address =. doi:10.1145/3610978.3640637 , url =

    arXiv 2024

  56. [56]

    International Joint Conferences on Artificial Intelligence (IJCAI) Survey Track , year =

    Yaru Niu and Zhenyu Jiang and Tianyi Zhou and Yanjie Ze and Pengfei Li and Yunchu Zhang and Pieter Abbeel and Sergey Levine and Chelsea Finn , title =. International Joint Conferences on Artificial Intelligence (IJCAI) Survey Track , year =

  57. [57]

    Fairness and Bias in Robot Learning , journal =

    Silja V. Fairness and Bias in Robot Learning , journal =. 2024 , doi =

    2024

  58. [58]

    Robotics and Autonomous Systems , year =

    Jonathan Fink and others , title =. Robotics and Autonomous Systems , year =

  59. [59]

    Nature Machine Intelligence , year =

    Sami Haddadin and others , title =. Nature Machine Intelligence , year =

  60. [60]

    2023 , doi =

    Artificial Intelligence Risk Management Framework (AI RMF 1.0) , institution =. 2023 , doi =

    2023

  61. [61]

    CAAI Artificial Intelligence Research , volume =

    Fuchun Sun and Runfa Chen and Tianying Ji and Yu Luo and Huaidong Zhou and Huaping Liu , title =. CAAI Artificial Intelligence Research , volume =. 2024 , doi =

    2024

  62. [62]

    IEEE Access , volume =

    Kawaharazuka, Kento and Oh, Jihoon and Yamada, Jun and Posner, Ingmar and Zhu, Yuke , title =. IEEE Access , volume =. 2025 , doi =

    2025

  63. [63]

    IEEE Internet of Things Journal , year =

    Satyam Gaba and Krutiksinh Rana and Siva Sai and Vinay Chamola and Dusit Niyato , title =. IEEE Internet of Things Journal , year =

  64. [64]

    The Quest for AI Sovereignty, Transparency and Accountability: Official Outcome of the UN IGF Data and Artificial Intelligence Governance Coalition , editor =

    Rama Vedashree and Jameela Sahiba and Bhoomika Agarwal , title =. The Quest for AI Sovereignty, Transparency and Accountability: Official Outcome of the UN IGF Data and Artificial Intelligence Governance Coalition , editor =. 2026 , pages =. doi:10.1007/978-3-032-02762-7_7 , isbn =

    work page doi:10.1007/978-3-032-02762-7_7 2026

  65. [65]

    Riek and Don Howard , title =

    Laurel D. Riek and Don Howard , title =. AAAI Workshop on AI and Ethics , year =

  66. [66]

    2026 , month =

    Physical. 2026 , month =

    2026

  67. [67]

    2026 , month =

    Why governance is the new infrastructure for physical. 2026 , month =

    2026

  68. [68]

    2026 , month =

    When. 2026 , month =

    2026

  69. [69]

    2026 , month =

    Standard for Foundational Framework for Manufacturing Physical Artificial Intelligence (Manufacturing. 2026 , month =

    2026

  70. [70]

    American Scholarly Research Conference (ASRC) , volume =

    Shaikh Shofiullah , title =. American Scholarly Research Conference (ASRC) , volume =. 2025 , pages =. doi:10.63125/jwm2e278 , url =

    work page doi:10.63125/jwm2e278 2025

  71. [71]

    Review of Managerial Science , year =

    Author(s) , title =. Review of Managerial Science , year =. doi:DOI , url =

  72. [72]

    arXiv preprint arXiv:2304.02924 , year =

    Yingbo Li and Anamaria-Beatrice Spulber and Yucong Duan , title =. arXiv preprint arXiv:2304.02924 , year =

    Pith/arXiv arXiv

  73. [73]

    Barfoot and Philippe Beaudoin and Yoshua Bengio and Jeannette Bohg and Oliver Brock and Isabelle Depatie and Dieter Fox and Daniel E

    Nicholas Roy and Ingmar Posner and Tim D. Barfoot and Philippe Beaudoin and Yoshua Bengio and Jeannette Bohg and Oliver Brock and Isabelle Depatie and Dieter Fox and Daniel E. Koditschek and Tom. From Machine Learning to Robotics: Challenges and Opportunities for Embodied Intelligence , journal =. 2021 , eprint =

    2021

  74. [74]

    Design in the Age of Artificial Intelligence , booktitle =

    Rog. Design in the Age of Artificial Intelligence , booktitle =

  75. [75]

    The International Journal of Robotics Research , volume =

    Roya Firoozi and Johnathan Tucker and Stephen Tian and Anirudha Majumdar and Jiankai Sun and Weiyu Liu and Yuke Zhu and Shuran Song and Ashish Kapoor and Karol Hausman and Brian Ichter and Danny Driess and Jiajun Wu and Cewu Lu and Mac Schwager , title =. The International Journal of Robotics Research , volume =. 2025 , doi =

    2025

  76. [76]

    SmartBot , year =

    Yifan Chen and Mingjie Wei and Xuesong Wang and others , title =. SmartBot , year =

  77. [77]

    Banerjee and David J

    Atul Thakur and Krishnanand Kaipa and Ashis G. Banerjee and David J. Cappelleri and Venkat N. Krovi and Satyandra Gupta , title =. Journal of Computing and Information Science in Engineering , volume =. 2025 , doi =

    2025

  78. [78]

    Trends and Challenges in Robot Manipulation , journal =

    Aude Billard and Danica Kragi. Trends and Challenges in Robot Manipulation , journal =. 2019 , doi =

    2019

  79. [79]

    Journal of Cleaner Production , year =

    Henning Wilts and others , title =. Journal of Cleaner Production , year =

  80. [80]

    Minds and Machines , volume =

    Iason Gabriel , title =. Minds and Machines , volume =. 2020 , doi =

    2020

Showing first 80 references.