New reports from high-level policy groups and military ethics experts reveal a critical failure to protect human agency within AI-integrated decision frameworks.

Two distinct institutional assessments published in September 2026 converge on a singular, urgent conclusion: the human element within AI-enabled systems currently lacks the necessary safeguards to prevent profound ethical and operational failures. The Elders, an international organization of former heads of state, and military ethics experts observing US Army operations, both argue that existing oversight mechanisms are insufficient to manage the risks posed by rapid automation in both civilian and combat environments.

The Elders released a formal set of ten principles on September 21, 2026, characterizing the current state of global AI governance as a systemic failure by sovereign states to protect their citizens from private-sector dominance. The organization asserts that governments have effectively abdicated their duty to regulate AI, leaving essential decisions regarding public services, infrastructure, and policing to a small group of technology firms. These companies, including OpenAI, Anthropic, Google, Meta, Microsoft, DeepSeek, and Alibaba, operate with minimal oversight despite the profound societal impact of their products.

The Elders’ framework demands that leaders prioritize humanity over profit, maintain human control over lethal decisions, and foster international cooperation to establish safety standards. By framing AI governance as a fundamental government obligation rather than a corporate choice, the group aims to shift the burden of accountability back to elected officials. This approach necessitates that leaders possess the ability to examine, question, and halt any system that influences critical public outcomes.

Military operations have reached a similar inflection point, as evidenced by a September 18, 2026, conference hosted by Task & Purpose for US Army Europe and Africa. Military chaplains and ethicists examined the psychological and moral consequences for service members who participate in AI-assisted targeting processes where they lack meaningful control. The 25th Infantry Division has already integrated digital sensor-to-shooter artillery systems, moving AI adoption from theoretical testing to operational reality.

Jesse Kirkpatrick, a research professor at George Mason University, notes that the psychological architecture of modern combat is evolving faster than the pastoral care infrastructure intended to support soldiers. He warns that moral injury—a deep psychological harm caused by actions that violate personal beliefs—may become increasingly prevalent as troops are held responsible for lethal outcomes they did not fully direct. The disconnect between nominal responsibility and genuine human agency creates a dangerous vacuum in military command structures.

Effective governance requires more than policy statements; it demands the implementation of robust monitoring that tracks unintended effects and security weaknesses post-deployment. The Elders argue that restraint must be a core component of any system involving the use of force, rejecting the deployment of weapons that autonomously select their own targets. Without such constraints, the speed and complexity of AI-driven targeting may overwhelm the human capacity for judgment, reducing operators to mere rubber stamps for algorithmic outputs.

The current reliance on AI-generated data creates a significant risk of moral harm when operators are pressured by time or interface design. Experts suggest that the Army’s response must extend beyond chaplaincy to include specialized mental health support and comprehensive AI literacy for all personnel involved in targeting decisions. True accountability necessitates that humans retain the authority to pause systems and contest outputs, ensuring that technical performance never supersedes human ethical standards.

The broader significance of these findings lies in the urgent need for enforceable international laws and transparent institutional oversight. As AI systems become more deeply embedded in sensitive sectors, the ability to inspect and reverse automated decisions will become a defining feature of responsible leadership. Long-term protection for society depends on the ability of governments to bridge the gap between technical capability and human experience, ensuring that AI serves public interests rather than undermining them.

Future watchpoints include the potential for formal negotiations between the United States and China regarding international safety standards for AI. Stakeholders must monitor whether military and corporate entities will adopt independent assurance mechanisms to address conflicts of interest in system design. The ultimate test of these governance frameworks will be their ability to provide tangible, continuous improvement in response to real-world incidents and emerging dependencies.

At Global Business & Economics Journal, we contribute to the  future of business and technology with news,  innovative research, expert insights, and thought-provoking analysis.