Job losses, bias, and inequality: Artificial intelligence reshaping modern society

Artificial intelligence is advancing at an unprecedented pace, altering economies, governance, and cultural life across the globe. A new review argues that AI is now both a powerful driver of innovation and a disruptive force demanding urgent ethical and policy oversight.

The study, titled "The Impact of Artificial Intelligence on Modern Society” and published in AI, examines the wide-ranging influence of AI across healthcare, finance, manufacturing, transportation, education, and governance. Drawing on literature, policy reports, and industry surveys from 2015 to 2025, it provides a comprehensive narrative of how AI technologies are reshaping social and economic landscapes.

Expanding applications across critical sectors

AI is no longer limited to research labs or narrow commercial applications. In healthcare, it is improving diagnostics, with systems already surpassing human performance in detecting breast cancer and diabetic retinopathy. AI tools are being deployed to analyze medical records, predict disease progression, and enhance triage processes in radiology, raising hopes for faster and more accurate patient care. Yet, the digitization of medical devices and supply chains also introduces new cybersecurity vulnerabilities, forcing hospitals and regulators to confront fresh risks.

The finance industry has been among the earliest adopters, using AI in fraud detection, investment advice, risk management, and regulatory compliance. Reinforcement learning models are being tested for portfolio optimization, while deep learning approaches outperform traditional models in predicting systemic risk. At the micro level, AI tools are also influencing trading strategies through sentiment analysis of financial news, blending traditional data with real-time market signals.

In education, researchers are advocating for AI literacy at early stages of schooling. The study emphasizes that integrating AI into curricula is not only about technical skills but also about fostering ethical awareness and critical thinking. AI-driven personalized learning systems promise tailored educational experiences, but concerns about teacher deskilling and bias in algorithmic assessment remain.

Manufacturing, too, illustrates the uneven reach of AI. Large firms deploy advanced AI-driven predictive maintenance and automation, but small and medium-sized manufacturers often lack affordable and user-friendly solutions. Proposed frameworks such as AI-assisted machine supervision aim to bridge this gap, providing real-time monitoring and anomaly detection to improve efficiency without prohibitive costs.

Transportation faces perhaps the most dramatic disruption on the horizon. Electric autonomous vehicles are positioned as a revolution in mobility, with the potential to upend traditional auto industries, urban planning, and energy systems. However, the study stresses that regulatory standards, infrastructure, and social acceptance remain far from ready.

Societal consequences: Employment, inequality and governance

Estimates suggest that around 40 percent of global jobs are exposed to automation. Routine and codifiable work faces the highest risk, while roles requiring creativity, empathy, or strategic judgment appear more resilient. The OECD projects that 14 percent of jobs are highly automatable and another 32 percent will undergo major changes. While new job categories may emerge, the pace of reskilling often lags behind technological progress.

The rise of gig platforms highlights AI’s double-edged role in labor markets. Matching algorithms set pay and allocate work, but lack transparency and may reinforce inequality. The data asymmetry between platforms and workers further compounds this imbalance, raising questions of fairness and accountability. Proposals for intelligent end-user assistants aim to empower workers, yet implementation remains limited.

Governance is another area where AI’s promise collides with risk. Algorithmic decision-making could optimize resource allocation and service delivery, but opaque models and bias undermine trust in public institutions. The review finds that political representation can be damaged when automated systems operate without transparency or adequate oversight. Comparative policy analysis shows sharp differences between global regions: the European Union emphasizes ethical safeguards, the United States prioritizes market-driven innovation, and China integrates AI into centralized planning and surveillance.

At the international level, the lack of harmonized standards risks fragmentation and regulatory arbitrage. Calls for global cooperation on AI governance are growing, particularly to address human rights concerns in authoritarian contexts where AI is used for surveillance and social control.

Ethical environmental, and security challenges

The study warns against over-reliance on opaque systems that may embed social biases, exacerbate inequality, or be misused for surveillance. Ethical debates increasingly center on privacy, accountability, and fairness, with “ethics by design” proposed as a guiding principle for future systems. AI is framed as a political technology that reflects the interests of its designers and users, rather than a neutral tool.

Environmental impacts are another critical concern. Training and operating large AI models consume vast amounts of energy and resources, raising questions about sustainability. Data centers, GPUs, and TPUs all contribute to rising carbon footprints, challenging the narrative of AI as a purely beneficial force. Proposals for sustainable practices, energy-efficient architectures, and green AI principles are gaining momentum, though implementation is uneven.

Security represents both a benefit and a risk. AI-enabled cybersecurity tools are increasingly used to detect and respond to attacks, offering faster identification of threats across networks and critical infrastructures. Yet the same technologies can be exploited by adversaries, leading to an escalating battle between offensive and defensive AI. Surveillance technologies further complicate the landscape, with citizens often trading privacy for convenience while corporations and states gather unprecedented amounts of personal data. Cross-cultural studies suggest that attitudes toward these developments vary: in Japan, AI is often associated with harmony, while in the West it frequently evokes dystopian fears.