Abstract
Inspired by the recent development of autonomous artificial intelligence (AI) systems in military and medical applications I envision the use of one such system, an AI empowered exoskeleton smart-suit called the Praetor Suit, to question the important ethical issues stemming from its use. The Praetor Suit would have the ability to monitor the service member’s physiological and psychological state, report that state to medical experts surveilling its operation through teleoperation and autonomously administer medical treatments based upon its evaluations. Doing so, it would effectively enhance the user’s operational capacity on the military mission field. The important ethical issues which are engaged and stem from the suit’s Monitor, Evaluation and Administration AI modules are primarily connected with issues of data privacy, human and AI autonomy, transparent automation processes and automation biases, AI explainability and trust. Lastly, in light of the medical automation worry, a positive portrayal of military human-AI partnership is given through the framework of human-AI symbiosis to engage the question does the introduction of the Praetor Suit inadvertently change the military role of the future medic from a purely non-combative to that of a hybrid or combative one.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
The Russian military rushes forward to develop their own third-generation Ratnik 3 suit which aims to integrate different important systems of life support, enhanced communication, protection and movement enhancement into their smart warrior suits (Ritsick 2018)
- 2.
The Praetor Suit is the armored suit worn by the Doom Slayer character in the popular game Doom (published in 2016). The suit is given to the player at the very beginning and is worn for the entirety of the game. Praetor Suit covers the Doom Marine’s whole body, including his arms. The suit is described as being made from nearly impenetrable material, and may be responsible for what appears to be the Doom Marine’s superhuman abilities (Praetor Suit 2018).
- 3.
The possibility of medical error is something which cannot be excluded from the medical profession as errors in medical diagnosis and treatment application are, unfortunately, a fact of medical life. Both humans and AI could err in the field of medical work, even though the reasons why the err may differ and differ drastically. For instance, even though the AI may never get tired or emotionally upset and may operate optimally and constantly without rest it, at least for now, has no capacities to improvise or adapt to unforeseen situations if it becomes necessary for the success of the medical operation.
- 4.
I stand inspired by Klein’s patient-machine-medic relation (Klein 2015) with the noted difference I purposefuly dislike using the term “machine” when describing human-AI relations as this term, engrained in popular Western culture, tends to produce discomforting connotations which tend to negatively impact the attitude towards the formation of human-AI relations. (Coeckelbergh 2014)
- 5.
AI Bias is recognized by leading institutions and experts (AI and bias – IBM Research – US 2018; Campolo et al. 2017) as one of the two biggest issues (the other problem is the “black box” or “explainable AI” issue) hindering further development and implementation of autonomous AI systems in the foreseeable future.
- 6.
“In other words, commanders do have the legal right to require service members to undergo certain medical procedures such as vaccinations and periodic medical examinations for fitness of duty.” (McManus et al. 2005, 1124)
- 7.
Such pre-mission obtained consent could be attained transparently and fully if, for instance, the suit’s operation would be field tested either in virtual space (through VR simulation) or real space. Such tests would not only psychologically accommodate the user to the suit’s use but would also fulfill the ethical necessity of informing the user on the suit’s beneficial operational capacities and the possible harmful consequences resulting from its use.
- 8.
Although there are different types of system failures which could result in such harmful consequences, it is paramount that they are not produced from design or system administration incompetence which results in system vulnerabilities or failures. Unfortunately, there will always exist a type of rare and unpredictable high impact events, “Black Swans” (Taleb and Chandler 2007), which cannot ever be fully excluded from manifesting even with best possible system design.
- 9.
This is especially important if the suit could have the capacity to administer or provide the user with stimulants or other enhancement drugs such as “stimulants to attenuate the effects of sleep loss (often referred to as ‘go pills’) and hypnotics to improve the ability to sleep (‘no-go pills’).” (Liivoja 2017, 5). It is paramount that such (combat) enhancement is not done autonomously without the knowledge of the user. In this regard one might even remember the 2001’s Space Odyssey HAL 9000, where HAL’s secret decision for the missions’ success results in an ethical disaster and loss of human life.
- 10.
Such a scenario also raises the question, are AIs then included into the chain of, medical, command?
- 11.
This could be achieved by the suit taking over or restricting movement autonomy and removing the service member from the harm’s way (or from inflicting harm to oneself or others), or, if necessary, by administering simple sedatives.
- 12.
In gaming worlds, the „DPS“ abbreviation is a colloquial term used to design a wide variety of player classes, playing styles or character builds aimed to fulfill a single specific goal – to deal as much damage as possible to the enemy. As such, those players who build their characters to become damage dealers usually forego all other character traits, for instance strong defense, in order to ensure their characters maximum offensive power.
- 13.
It is often the case how players have to pass through the same mission more than once as combat usually occurs in stages with different enemy behavior or environment rules occurring for each separate stage. For this reason, experienced medic players are highly sought for especially by inexperienced, or first-time, players venturing into the same mission.
- 14.
In this regard, I agree with Liivoja’s (2017) lucid analysis of the reasons why medical personnel engaged in the biomedical enhancement of soldiers would suffer a “loss of special protection that such personnel and units enjoy” (Liivoja 2017, 27). Still, I take that the prospect of AI empowered enhancement has the potential to generate more fundamental changes to the role and purpose of the military medic than the ones exemplified in cases of biomedical enhancement.
References
Acemoglu, Daron, and Pascual Restrepo. 2018. Artificial intelligence, automation and work. Cambridge, MA: National Bureau of Economic Research. Ssrn. https://doi.org/10.2139/ssrn.3098384.
AI and bias – IBM Research – US. 2018. Research.ibm.com
Agrawal, Ajay, Joshua S. Gans, and Avi Goldfarb. 2018. Exploring the Impact of Artificial Intelligence: Prediction Versus Judgment.. Rotman School of Management, https://doi.org/10.2139/ssrn.3177467.
Angwin, Julia, and Surya Mattu. 2018. Machine Bias — ProPublica. ProPublica.
Azevedo, Carlos R.B., Klaus Raizer, and Ricardo Souza. 2017. A vision for human-machine mutual understanding, trust establishment, and collaboration. 2017 IEEE conference on cognitive and computational aspects of situation management, CogSIMA 2017, 9–11. https://doi.org/10.1109/COGSIMA.2017.7929606.
Balkin, Jack M. 2017. Free speech in the algorithmic society: Big data, private governance, and new school speech regulation. SSRN Electronic Journal. Elsevier BV. https://doi.org/10.2139/ssrn.3038939.
Bennett, C.C., and K. Hauser. 2013. Artificial intelligence framework for simulating clinical decision-making: A Markov decision process approach. Artificial Intelligence in Medicine 57: 9–19. https://doi.org/10.1016/j.artmed.2012.12.003.
Brynjolfsson, Erik, and Tom Mitchell. 2017. What can machine learning do? Workforce implications: Profound change is coming, but roles for humans remain. Science 358: 1530–1534. https://doi.org/10.1126/science.aap8062.
Casner, Stephen M., Edwin L. Hutchins, and Don Norman. 2016. The challenges of partially automated driving. Communications of the ACM 59: 70–77. https://doi.org/10.1145/2830565.
Coeckelbergh, Mark. 2014. The moral standing of machines: Towards a relational and non-Cartesian moral hermeneutics. Philosophy and Technology 27: 61–77. https://doi.org/10.1007/s13347-013-0133-8.
Cohn, Jeffrey F., Tomas Simon Kruez, Iain Matthews, Ying Yang, Minh Hoai Nguyen, Margara Tejera Padilla, Feng Zhou, and Fernando De La Torre. 2009. Detecting depression from facial actions and vocal prosody. Proceedings – 2009 3rd international conference on affective computing and intelligent interaction and workshops, ACII 2009. https://doi.org/10.1109/ACII.2009.5349358.
Campolo, Alex, Madelyn Sanfilippo, Meredith Whittaker, and Kate Crawford. 2017. AI Now 2017 Report. https://ainowinstitute.org/AI_Now_2017_Report.pdf.
Dietvorst, Berkeley J., Joseph P. Simmons, and Cade Massey. 2014. Algorithm aversion: People erroneously avoid algorithms after seeing them err. SSRN Electronic Journal. Elsevier BV. https://doi.org/10.2139/ssrn.2466040.
EU Declaration on Cooperation on Artificial Intelligence – JRC Science Hub Communities – European Commission. 2018. JRC Science Hub Communities.
Fischer, Alastair J., and Gemma Ghelardi. 2016. The precautionary principle, evidence-based medicine, and decision theory in public health evaluation. Frontiers in Public Health 4: 1–7. https://doi.org/10.3389/fpubh.2016.00107.
Gao, Wei, Sam Emaminejad, Hnin Yin Yin Nyein, Samyuktha Challa, Kevin Chen, Austin Peck, Hossain M. Fahad, Hiroki Ota, Hiroshi Shiraki, Daisuke Kiriya, Der-Hsien Lien, George A. Brooks, Ronald W. Davis, and Ali Javey. 2016. Fully integrated wearable sensor arrays for multiplexed in situ perspiration analysis. Nature 529 (7587): 509–514. https://doi.org/10.1038/nature16521.
Gunning, David. 2017. Explainable artificial intelligence (xai). Defense Advanced Research Projects Agency (DARPA).
Hengstler, Monika, Ellen Enkel, and Selina Duelli. 2016. Technological forecasting & social change applied arti fi cial intelligence and trust — The case of autonomous vehicles and medical assistance devices. Technological Forecasting and Social Change 105: 105–120. https://doi.org/10.1016/j.techfore.2015.12.014.
Lawless, W.F., Ranjeev Mittu, Stephen Russell, and Donald Sofge. 2017. Autonomy and artificial intelligence: A threat or savior? Cham: Springer. https://doi.org/10.1007/978-3-319-59719-5.
Lee, Chien-Cheng, Cheng-Yuan Shih, Wen-Ping Lai, and Po-Chiang Lin. 2012. An improved boosting algorithm and its application to facial emotion recognition. Journal of Ambient Intelligence and Humanized Computing 3: 11–17. https://doi.org/10.1007/s12652-011-0085-8.
Liivoja, Rain. 2017. Biomedical enhancement of warfighters and the legal protection of military medical personnel in armed conflict. Medical Law Review 0: 1–28. https://doi.org/10.1093/medlaw/fwx046.
Kalantar-zadeh, Kourosh, Nam Ha, Jian Zhen Ou, and Kyle J. Berean. 2017. Ingestible Sensors. ACS Sensors 2 (4): 468–483. https://doi.org/10.1021/acssensors.7b00045.
Klein, Eran. 2015. Models of the Patient-Machine-Clinician Relationship in Closed-Loop Machine Neuromodulation. In Machine Medical Ethics. Intelligent Systems, Control and Automation: Science and Engineering, ed. S. van Rysewyk and M. Pontier, vol. 74. Cham: Springer. https://doi.org/10.1007/978-3-319-08108-3_17.
McManus, John, Sumeru G. Mehta, Annette R. McClinton, Robert A. De Lorenzo, and Toney W. Baskin. 2005. Informed consent and ethical issues in military medical research. Academic Emergency Medicine 12: 1120–1126. https://doi.org/10.1197/j.aem.2005.05.037.
Neumann, Peter G. 2016. Risks of automation. Communications of the ACM 59: 26–30. Association for Computing Machinery (ACM). https://doi.org/10.1145/2988445.
Ng, Andrew. 2017. Andrew Ng: Artificial intelligence is the new electricity. YouTube.
Praetor Suit. 2018. Doom Wiki.
Rajpurkar, Pranav, Jeremy Irvin, Robyn L. Ball, Kaylie Zhu, Brandon Yang, Hershel Mehta, and Tony Duan et al. 2018. Deep Learning For Chest Radiograph Diagnosis: A Retrospective Comparison Of The Chexnext Algorithm To Practicing Radiologists. PLOS Medicine 15 (11): e1002686. doi:https://doi.org/10.1371/journal.pmed.1002686.
Reidsma, D., V. Charisi, D.P. Davison, F.M. Wijnen, J. van der Meij, V. Evers, et al. 2016. The EASEL project: Towards educational human-robot symbiotic interaction. In Proceedings of the 5th international conference on living machines, Lecture notes in computer science; Vol. 9793, ed. N.F. Lepora, A. Mura, M. Mangan, P.F.M.J. Verschure, M. Desmulliez, and T.J. Prescott, 297–306. Cham: Springer International Publishing. https://doi.org/10.1007/978-3-319-42417-0_27.
Ritsick, Colin. 2018. Ratnik 3 – Russian Combat Suit | Future infantry exoskeleton combat system. Military Machine.
Rosenthal, Stephanie, J. Biswas, and M. Veloso. 2010. An effective personal mobile robot agent through symbiotic human-robot interaction. Proceedings of the 9th international conference on autonomous agents and multiagent systems (AAMAS 2010), 915–922.
Ross, Casey, and Ike Swetlitz. 2017. IBM pitched Watson as a revolution in cancer care. It’s nowhere close. STAT.
Sim, Jai Kyoung, Sunghyun Yoon, and Young-Ho Cho. 2018. Wearable sweat rate sensors for human thermal comfort monitoring. Scientific Reports, 8. Springer Nature. https://doi.org/10.1038/s41598-018-19239-8.
Taleb, Nassim, and David Chandler. 2007. The black swan. Rearsby: W.F. Howes.
Tavani, Herman T. 2015. Levels of trust in the context of machine ethics. Philosophy and Technology 28: 75–90. https://doi.org/10.1007/s13347-014-0165-8.
Taylor, Earl L. 2017. Making sense of “algorithm aversion”. Research World 2017: 57. Wiley. https://doi.org/10.1002/rwm3.20528.
The United States Special Operations Command. 2018. SOFIC 2018 conference program & exhibits guide 2018. In 2018 United States special operations forces industry conference (SOFIC) and exhibition. https://www.sofic.org/-media/sites/sofic/documents/sofic_2018_final-low-res3.ashx.
Warrick, Philip A, and Masun Nabhan Homsi. 2018. Ensembling convolutional and long short-term memory networks for electrocardiogram arrhythmia detection. Physiological Measurement. IOP Publishing. https://doi.org/10.1088/1361-6579/aad386.
White, Stephen. 2008. Brave new world: Neurowarfare and the limits of international humanitarian law. Cornell International Law Journal 41 (1): 177–210.
Wortham, Robert H., Andreas Theodorou, and Joanna J. Bryson. 2016. What does the robot think? Transparency as a fundamental design requirement for intelligent systems. IJCAI-2016 ethics for artificial intelligence workshop.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2020 Springer Nature Switzerland AG
About this chapter
Cite this chapter
Miletić, T. (2020). Military Medical Enhancement and Autonomous AI Systems: Requirements, Implications, Concerns. In: Messelken, D., Winkler, D. (eds) Ethics of Medical Innovation, Experimentation, and Enhancement in Military and Humanitarian Contexts. Military and Humanitarian Health Ethics. Springer, Cham. https://doi.org/10.1007/978-3-030-36319-2_11
Download citation
DOI: https://doi.org/10.1007/978-3-030-36319-2_11
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-36318-5
Online ISBN: 978-3-030-36319-2
eBook Packages: Religion and PhilosophyPhilosophy and Religion (R0)