Ethical Considerations in Defense Technology Development

✈️ 🇪🇺 « Trustworthy AI in Defence »: The European Way 🗞️The European Defence Agency’s White Paper is out! At a time when global powers are racing to develop & deploy AI-enabled defence capabilities,the European way =tech innovation + ethical responsibility, operational effectiveness + legal compliance, strategic autonomy + respect for human dignity & democratic values. 🔹AI in defence as legally compliant, ethically sound, technically robust, societally acceptable. 1 🤝🏻Principles of Trustworthiness 🔹foundational principles for trustworthy AI in defence: accountability, reliability, transparency, explainability, fairness, privacy, human oversight. Not optional but integral to the legitimacy of AI systems used by European armed forces. 2. Ethical and Legal Compliance 🔹 Europe’s commitment is to effective military capabilities but also to a rules-based international order. The EU explicitly rejects the idea that technological advancement justifies the erosion of ethical norms. 🔹 importance of ethical review mechanisms, institutional safeguards, alignment with #EU legal frameworks=a legal-ethical backbone ensuring trustworthiness is a practical requirement embedded into every phase of AI development/deployment. 3. Risk Assessment & Mitigation 🔹 EU’s precautionary principle=>rigorous & ongoing risk assessments of AI systems, incl. risks related to technical failures, misuse, bias, and unintended escalation in operational contexts. To anticipate harm before it materializes and equip systems with built-in safeguards 🔹Risk mitigation not only a technical task but an ethical &strategic imperative in high-stakes domains (targeting, threat detection, autonomous mobility). 4. 👁️Human Oversight & Control 🔹The EU rejects fully autonomous weapon systems operating without human intervention in critical functions like the use of force. The Paper calls for clear human-in-the-loop models, where operators retain oversight, intervention capability, and accountability. = safeguards democratic accountability & operational reliability, ensuring no algorithm makes life-and-death decisions. 5. Transparency and Explainability 🔹transparent #AI systems, not black-box models : decision-making processes understandable by users & traceable by designers. Key for after-action reviews, audits, & compliance. Strong stance on explainability 6. European Cooperation &Standardization 🔹Enhanced cooperation and harmonization in defence AI : shared definitions, frameworks to ensure interoperability, avoid duplication, promote a common culture of responsibility. 🔹 joint work on certification processes, training, testing environments 7. Continuous Monitoring and Evaluation 🔹ongoing monitoring, validation, recalibration of AI tools throughout their deployment. «trustworthiness must be maintained, not assumed » =The European way: lead not by imitating others’ race toward automation at any cost, but by demonstrating security, innovation, and values can go hand in hand

𝗧𝗵𝗲 𝗦𝗶𝗹𝗲𝗻𝘁 𝗗𝗮𝗻𝗴𝗲𝗿 𝗼𝗳 𝗨𝗻𝗰𝗵𝗲𝗰𝗸𝗲𝗱 𝗔𝗜 𝗶𝗻 𝗠𝗶𝗹𝗶𝘁𝗮𝗿𝘆 𝗖𝗼𝗺𝗺𝗮𝗻𝗱 Battlefield complexity generates data faster driving risky AI shortcuts. Ukraine shows how fragile AI is when electronic warfare disrupts communications. Algorithms often impose harsher outcomes revealing serious bias. The risk is in our failure to govern AI responsibly. The Alan Turing Institute’s report sheds light on the high-stakes tension between AI’s promise and ethical responsibility in military command. 𝗧𝗵𝗲 𝗣𝗿𝗼𝗺𝗶𝘀𝗲 𝗼𝗳 𝗠𝗶𝗹𝗶𝘁𝗮𝗿𝘆 𝗔𝗜 • Speeds up data analysis for situational awareness • Supports commanders in handling overwhelming battlefield complexity • But warns against blind reliance on algorithms 𝗗𝗼𝗰𝘁𝗿𝗶𝗻𝗲 𝗮𝗻𝗱 𝘁𝗵𝗲 𝗖𝗼𝗺𝗯𝗮𝘁 𝗘𝘀𝘁𝗶𝗺𝗮𝘁𝗲 • The Seven-Step Combat Estimate Process helps shape mission plans • AI can suggest Courses of Action (COA) • Human judgment remains essential for proportionality and legality 🎯𝗖𝗵𝗮𝗹𝗹𝗲𝗻𝗴𝗲𝘀  Technology and Battlefield Noise • Electronic warfare can sever AI’s data links • Real-world conflicts (e.g., Ukraine) show AI’s vulnerability • Overreliance risks decision-making paralysis in contested environments Military Command and Its Responsibilities • Commanders are legally accountable for decisions • AI cannot replace moral reasoning • International Humanitarian Law (IHL) requires clear human control Organisational and Cultural Shifts • Private sector innovation clashes with military identity • Risk of opaque algorithms disempowering human judgment • Collapse of information ecosystems threatens resilience 𝗧𝗵𝗲 𝗪𝗮𝘆 𝗙𝗼𝗿𝘄𝗮𝗿𝗱 Center human judgment in AI deployment • Design AI as decision-support, not decision-maker • Build for contested, degraded environments • Establish clear accountability frameworks • Promote cross-sector collaboration for resilient ecosystems 𝗕𝗼𝘁𝘁𝗼𝗺 𝗟𝗶𝗻𝗲  The next evolution of warfare will depend on wiser collaboration and designing responsible AI that amplifies ethics, accountability and human insight at the heart of every decision. Prof. Dr. Ingrid Vasiliu-Feltes|Helen Yu|JOY CASE|Hr Dr. Takahisa Karita|Antonio Grasso|Nicolas Babin |Alberto Espinosa Machado|Dr. Ram Kumar|Phillip J Mostert| Sara Simmonds |Anthony Rochand|Prasanna Lohar|Shalini Rao #AI #EthicalAI #AIinMilitary #ResponsibleAI #DigitalTrust #AIGovernance #TechForGood #InclusiveInnovation

I’ve been reading the U.S. “Artificial Intelligence Strategy for the Department of War” memorandum and one section deserves serious attention: “Clarifying ‘Responsible AI’ - Out with Utopian Idealism, In with Hard-Nosed Realism.” The document explicitly states that AI models should be free from “ideological tuning” and “usage policy constraints” that may limit lawful military applications. The focus is speed, dominance, and unrestricted operational use. Let’s be honest: this is a very clear strategic signal. It prioritises acceleration and military advantage over guardrails, ethics frameworks, and broader accountability structures. “Speed wins” is repeated throughout the document. Risk is framed mainly as “moving too slowly,” not as misuse, escalation, or unintended consequences. From a deterrence and capability perspective, this is understandable. From a governance, alliance, and societal perspective, it raises difficult questions. If responsible AI is reduced to “objectivity benchmarks” and “any lawful use” clauses, where do transparency, bias mitigation, civilian harm prevention, and long-term stability sit? Who defines “objective”? Who audits it? Who is accountable when systems fail or are misused? This matters far beyond the U.S. American defence AI ecosystems are deeply connected to NATO, partner nations, and allied procurement chains. Standards set here will ripple across Europe and beyond. They will influence how systems are designed, exported, integrated, and trusted. At the same time, the strategy is largely silent on concrete mechanisms for ethical oversight, civilian protection, cross-border governance. It speaks extensively about infrastructure, data access, speed, competition but very little about how responsibility is operationalised in practice. We are seeing a shift from “trustworthy AI” to “AI dominance first.” That shift may bring short-term advantage. But without strong literacy, governance, shared rules, it also increases strategic risk. Especially in an era where autonomous systems, decision-support tools, AI-enabled targeting are becoming normalised. Responsible AI in defence cannot be reduced to a slogan or a procurement checkbox. It has to mean: – Clear accountability chains – Independent validation&testing – Shared allied standards – Robust legal&ethical review – Real investment in data, governance, and human oversight Otherwise, we are building systems that are powerful, fast and dangerously fragile. This is not about slowing innovation but about ensuring that speed does not outpace responsibility. In defence, the cost of getting this wrong is measured in lives not just failed pilots or sunk investments. This conversation is not optional anymore. It is central to security, alliances, and the future of warfare. #AIinDefence #ResponsibleAI #MilitaryAI #NATO #DefenceInnovation #AIgovernance https://lnkd.in/d_2Hy2dD

AI Ethics Collide with Military Strategy Introduction Artificial intelligence has rapidly become embedded in modern military operations. As governments seek technological advantage, AI firms face a difficult balance between commercial opportunity and ethical responsibility. A growing dispute between the U.S. Department of Defense and AI developer Anthropic illustrates how governance of powerful AI tools is becoming a strategic and legal battleground. Key Developments Military adoption of AI expands The U.S. military has accelerated adoption of generative AI tools to support planning, analysis, and operational decision-making. Large technology companies see government contracts as an important revenue source after investing heavily in computing infrastructure. Anthropic’s ethical restrictions Anthropic placed two key limits on how its Claude AI system may be used. The company prohibits use for fully autonomous weapons systems. It also restricts use for mass domestic surveillance of U.S. citizens. Pentagon response escalates conflict U.S. defense leadership argued these restrictions interfere with operational authority. Defense officials labeled Anthropic a supply-chain national security risk. Contractors using Claude were barred from working with the Department of Defense. Legal and industry reaction Anthropic responded with a federal lawsuit claiming retaliation for enforcing responsible AI safeguards. The company argues the government ban is overly broad and harms legitimate commercial relationships. At the same time, OpenAI clarified its own policies prohibiting intentional domestic surveillance of Americans. User migration and market impact The dispute triggered increased interest in Claude among users seeking stronger AI safety guardrails. The Pentagon indicated Claude may continue to be used temporarily during a transition period. The outcome of Anthropic’s legal challenge could influence future AI procurement policies. Conclusion: Why This Matters The confrontation highlights a defining tension of the AI era. Governments want unrestricted access to powerful tools for national security, while technology companies seek to impose ethical boundaries on how those tools are deployed. How this balance is resolved will shape the governance of AI in warfare, cybersecurity, and national defense for years to come.

🔫 Autonomous AI weapons are no longer theoretical — they’re being built, tested, and in some cases, deployed. We’re entering a new era where rifles can identify targets using facial recognition, track movement patterns, and make split-second decisions — all without a human finger on the trigger. This raises a critical question: If machines can be trained to recognize enemies... can they also be trusted not to misidentify civilians? On one hand, AI promises precision, speed, and reduced risk to human soldiers. On the other, it challenges the very foundation of accountability in warfare. If an autonomous weapon makes a fatal mistake, who is responsible? The developer? The military commander? The algorithm? It’s worth noting that even landmines — passive, outdated tech — are a form of autonomous weapon, indiscriminate and still killing decades after wars end. So if we’re already living with that level of automation, could smarter AI weapons offer a more ethical path forward? The debate is no longer about if these systems should exist — it’s about how they are governed. Let’s not wait for the technology to outpace our values. 🦾 #RoboSapiens #MilTech #Defense #AutonomousWeapons #AIinWarfare #MilitaryTech #DefenseEthics #ArtificialIntelligence #ResponsibleAI #FutureOfWar

AI occupies a unique position in terms of dual-use technologies (DUT), reflecting its potential for both beneficial applications and military utilisation. AI's dual-use nature poses significant regulatory and ethical challenges, notably in its military dimensions which remain largely outside the ambit of civilian legislation such as the proposed AI Act. DUT are those with potential applications in both civilian and military domains. The essence of DUT lies in its versatility; the same technology that propels advancements in healthcare, education, and industry can also be adapted for surveillance, autonomous weaponry, and cyber warfare. This inherent ambiguity in application makes the governance of DUT, especially AI, a complex task. The AI Act primarily addresses civilian uses of AI, focusing on ethical guidelines, data protection, and transparency. Military applications of AI, by contrast, remain largely outside the scope of this act and other similar legislative efforts globally. The nuanced aspect of dual-use capabilities in AI brings software contracts into focus, serving as a critical instrument in governing the use, deployment, and development of AI technologies. Software contracts between developers, vendors, and users sometimes contain dual-use provisions to explicitly govern the use of the technology in both civilian and military contexts. These provisions are designed to ensure that the deployment of AI technologies aligns with legal standards, ethical norms, and, when applicable, international regulations. Dual-use clauses in software contracts may include restrictions on usage, export controls, compliance with international law, and requirements for end-use monitoring. Restrictions on Usage: Contracts may specify permissible uses of the software, explicitly prohibiting or restricting its application in military settings without proper authorisation. This helps in mitigating the risks associated with unintended or unauthorised military use of AI technologies. Export Controls: Given the potential military applications of AI, software contracts often include clauses related to export controls, requiring compliance with national and international regulations governing the export of dual-use technologies. This ensures that AI technologies do not inadvertently contribute to proliferation or escalate geopolitical tensions. Compliance with International Law: Provisions may also require that the use of AI technologies, particularly in military contexts, complies with international humanitarian law and other relevant legal frameworks. This is crucial in ensuring that the deployment of AI in warfare adheres to principles of distinction, proportionality, and necessity. It is clear that addressing the dual-use dilemma of AI extends beyond contractual measures. It requires a holistic approach that combines legal frameworks, ethical considerations, and international cooperation.

The U.S. Department of Defense just announced formal partnerships with six leading AI labs: Anthropic, Cohere, Meta, Microsoft, OpenAI, and Google DeepMind. The purpose? To promote what it calls the “safe, responsible, and ethical” use of AI in the military domain. That phrase “responsible military AI” deserves more scrutiny than it’s getting. Because we’re not talking about edge-case automation here. We’re talking about foundation models: systems trained on vast public corpora, originally justified as general-purpose tools for language, vision, reasoning, and creativity. And now, they’re being integrated into defence workflows. This isn’t a fringe development. It’s a structural pivot - from openness to strategic entrenchment. From civilian infrastructure to military-grade capabilities. And the shift is being wrapped in the same vocabulary of responsibility, safety, and alignment that was originally designed to signal restraint. But responsibility without transparency isn’t ethics. It’s branding. The announcement gestures at “managing risks,” but offers no detail on what those risks are, who defines them, or how they’ll be governed. In that vacuum, responsibility becomes a posture, more about reassurance than reflection. When labs talk about “ethical military use” without public definitions, enforceable constraints, or independent oversight, what they’re really offering is ambiguity as policy. The ethical language doesn’t constrain the activity; it legitimises it. It functions as a shield: a way to reframe risk as leadership, and moral complexity as operational necessity. And that matters, because these systems were trained on publicly available data, developed using civilian research infrastructure, and marketed as tools for universal benefit. Their capabilities were cultivated under the banner of progress. Now they are being repurposed for warfare. That doesn’t automatically make it wrong. But it does make it urgent. Urgent to ask what we mean by “alignment” when it applies to both democratic ideals and battlefield decisions. Urgent to interrogate the incentives that drive companies to warn of existential threat one month and partner with militaries the next. There is nothing wrong with national security partnerships per se. But there is something deeply dangerous about the fusion of ethical language and strategic opacity, especially when the consequences are highest. If we’re going to allow AI to shape military infrastructure, then the burden is not just to develop responsibly, but to govern visibly. Not just to declare ethics, but to embody constraint. And not just to promise alignment, but to decide - publicly - what exactly we are aligning to. Because if we fail to do that, then the phrase “responsible AI” becomes exactly what critics fear: a beautifully worded mask for a power structure that no longer bothers to explain itself. The silence around this topic is more dangerous than any discomfort I feel in raising it.

✳ Bridging Ethics and Operations in AI Systems✳ Governance for AI systems needs to balance operational goals with ethical considerations. #ISO5339 and #ISO24368 provide practical tools for embedding ethics into the development and management of AI systems. ➡Connecting ISO5339 to Ethical Operations  ISO5339 offers detailed guidance for integrating ethical principles into AI workflows. It focuses on creating systems that are responsive to the people and communities they affect. 1. Engaging Stakeholders  Stakeholders impacted by AI systems often bring perspectives that developers may overlook. ISO5339 emphasizes working with users, affected communities, and industry partners to uncover potential risks and ensure systems are designed with real-world impact in mind. 2. Ensuring Transparency  AI systems must be explainable to maintain trust. ISO5339 recommends designing systems that can communicate how decisions are made in a way that non-technical users can understand. This is especially critical in areas where decisions directly affect lives, such as healthcare or hiring. 3. Evaluating Bias  Bias in AI systems often arises from incomplete data or unintended algorithmic behaviors. ISO5339 supports ongoing evaluations to identify and address these issues during development and deployment, reducing the likelihood of harm. ➡Expanding on Ethics with ISO24368  ISO24368 provides a broader view of the societal and ethical challenges of AI, offering additional guidance for long-term accountability and fairness. ✅Fairness: AI systems can unintentionally reinforce existing inequalities. ISO24368 emphasizes assessing decisions to prevent discriminatory impacts and to align outcomes with social expectations.  ✅Transparency: Systems that operate without clarity risk losing user trust. ISO24368 highlights the importance of creating processes where decision-making paths are fully traceable and understandable.  ✅Human Accountability: Decisions made by AI should remain subject to human review. ISO24368 stresses the need for mechanisms that allow organizations to take responsibility for outcomes and override decisions when necessary. ➡Applying These Standards in Practice  Ethical considerations cannot be separated from operational processes. ISO24368 encourages organizations to incorporate ethical reviews and risk assessments at each stage of the AI lifecycle. ISO5339 focuses on embedding these principles during system design, ensuring that ethics is part of both the foundation and the long-term management of AI systems. ➡Lessons from #EthicalMachines  In "Ethical Machines", Reid Blackman, Ph.D. highlights the importance of making ethics practical. He argues for actionable frameworks that ensure AI systems are designed to meet societal expectations and business goals. Blackman’s focus on stakeholder input, decision transparency, and accountability closely aligns with the goals of ISO5339 and ISO24368, providing a clear way forward for organizations.

The recent development of a “dual-loop” non-invasive brain-computer interface (BCI) system by researchers at Tianjin University and Tsinghua University represents a significant advancement in reciprocal human-machine learning (see: https://lnkd.in/eDrdCF7B). The system, which has demonstrated real-time control of a drone, exemplifies rapid progress in neurotechnology, and while the stated intention is for research and clinical applications, such innovation also raises critical dual-use, neuroethical concerns that must be addressed. Dual-use technologies are those that can be utilized for both beneficial and potentially harmful purposes. The “dual-loop” BCI system, designed to enhance human-machine interactions, holds promise for augmenting human capabilities, which could be purposed for military applications, such as controlling unmanned systems or optimizing warfighter and intelligence operator performance as Rachel Wurzman and I noted some years ago in the journal STEPS (#STEPS). More broadly, this type of BCI system could be employed in other occupational settings to evaluate and affect cognitive capabilities and quality and extent of work output. If viewed through a relatively optimistic lens, this could be seen as positively valent. But this prompts questions of equity and access: such use may exacerbate social inequalities if access is limited to certain groups and widen the divide between those with enhanced capabilities and those without. Moreover, integration of such BCIs into daily life prompts several ethical questions about privacy and consent – namely unauthorized or mandatory monitoring – and influence -of an individual’s cognitive and behavioral patterns. Such engagement can be used to direct neurocognitive processes, with defined risk of controlling individual agency, and diminishing personal autonomy. And as with any emerging technology the longterm use of such a BCI system remains uncertain. To navigate these dual-use, neuroethical challenges, a multifaceted approach is recommended that entails (1) international collaboration – or at least cooperation – to establishing global standards and agreements to regulate responsible development and application of BCI technologies; (2) developing comprehensive ethical guidelines, informed by diverse multinational stakeholders to inform responsible innovation and use; (3) public engagement to enable more informed social awareness and attitudes; and (4) continuous oversight of these cooperatives to monitor – and course correct - BCI research and applications. Thus, while this “dual-loop” non-invasive BCI system offers promising advancements in human-machine interaction, it is imperative to address the associated dual-use and neuroethical issues. Proactive and collaborative efforts are essential to harness the benefits of such technologies while mitigating their potential risks. #dual loop #BCI #dual use #Neurotechnology #neuroethics