Artificial Intelligence "On the Front Lines": Big Tech Under Pressure Between Ethics, Smartphones, and Military Software

In recent weeks, the topic of artificial intelligence applied to military and national security contexts has reached a boiling point. Major technology companies — from Google to Amazon, from Microsoft to Palantir — find themselves at the center of a heated debate that touches simultaneously on international politics, innovation ethics, and the future of AI software. It's no longer a theoretical question: AI is literally "on the front lines," integrated into surveillance systems, intelligence analysis, autonomous drones, and command infrastructure.

But what happens when the same platforms that power your Netflix recommendations or searches on your smartphone are repurposed to guide military operations? And most importantly: who decides where to draw the line between legitimate innovation and complicity in armed conflicts? These questions are splitting Silicon Valley and challenging the entire "tech for good" narrative that Big Tech has built over the past twenty years.

Italy, like the rest of Europe, watches these developments with growing attention, also because the implications concern not only global geopolitics but also investment choices, AI regulation, and adoption across our continent. Understanding what is happening is essential for anyone who wants to navigate the technological landscape of 2026.

Military AI Is Not Science Fiction: Here's What's Really Happening

A few years ago, the idea of an algorithm that helped identify enemy targets or manage weapons systems was confined to Tom Clancy novels. Today it is operational reality. Several government programs — primarily in the United States, but also in Israel, China, and Russia — use artificial intelligence software to process enormous volumes of data in real time: satellite images, intercepted communications, troop movements, predictive analysis.

The Pentagon's Maven program, launched back in 2017, was the first major case to make headlines when Google, after internal employee protests, chose not to renew the contract. Since then, however, things have changed dramatically. In 2026, Google has signed new agreements with government defense agencies through more opaque corporate structures. Microsoft has integrated advanced AI capabilities into its JEDI contract with the US Department of Defense. Amazon Web Services hosts critical infrastructure for NATO.

The numbers speak clearly:

• The global market for AI in defense was worth approximately $38 billion in 2024 and is estimated to reach $90 billion by 2030
• The United States alone allocated over $2.5 billion in AI-related defense contracts in 2025 alone
• Palantir, among the most exposed companies in this sector, has seen its stock price grow by 140% in 18 months, driven precisely by military contracts
• AI is being used in at least 37 active conflicts worldwide, according to a report from the International Committee for Robot Arms Control

All this happens while citizens use the same smartphones on which "civilian" versions of these systems run, and while the boundary between commercial software and military software becomes increasingly blurred.

Big Tech in the Crosshairs: Between Billion-Dollar Contracts and Internal Pressure

Big Tech is under pressure from two opposite and simultaneous fronts. On one hand, governments — especially the American one — exert enormous pressure for technology companies to contribute to national technological supremacy. On the other, employees, shareholders, and the public demand transparency and clear ethical boundaries.

The most recent case involves Google DeepMind, which in 2025 signed an agreement with the British Ministry of Defence for predictive analysis in military logistics. When news of the deal leaked, hundreds of employees signed internal petitions, and some leading researchers resigned. The incident reignited the debate over how transparent AI companies should be regarding the use of their models.

Big Tech faces increasingly difficult choices:

1. Accept lucrative government contracts risking alienating talent and damaging their ethical reputation
2. Refuse military contracts losing billions of dollars and leaving space for less scrupulous competitors
3. Create separate divisions for defense contracts, as Microsoft did with its "Government Security" team — a solution many define as "ethics theater"

Making the picture even more complicated is the question of large language models (LLMs): generative AI systems like GPT-5 or Gemini Ultra can be adapted for military applications with a surprisingly limited amount of fine-tuning. This means that the same technologies powering voice assistants on your smartphones can, in principle, be repurposed in intelligence or surveillance contexts.

Smartphones, Software, and Dual-Use AI: The Case of the End Consumer

The question of "dual use" — both civilian and military use of the same technology — concerns not just large servers and data centers. It also concerns the devices that millions of Italians carry in their pockets every day.

Modern high-end smartphones are equipped with neural chips (like Google's Tensor G4 or Apple's A18 Pro chip) designed to run artificial intelligence models directly on the device, without sending data to the cloud. This "on-device AI" capability was developed primarily for consumer applications — computational photography, real-time translation, voice assistants. But the same chips and the same software can be used for facial recognition, biometric analysis, and surveillance applications.

Here are some concrete examples of how dual-use AI manifests in daily life:

• Facial recognition: present on every modern smartphone to unlock the device, but the same technology is used in mass surveillance systems in China and — in more limited forms — in various European cities
• Image analysis: AI photo apps can now identify objects, people, and places with military precision — it's no accident that some of these features are being studied with interest by intelligence services
• Translation and NLP: language models built into smartphones are now capable of translating rare dialects and detecting communication nuances useful in intelligence contexts

The regulatory problem is enormous. The AI Act, which entered fully into force in 2025, prohibits certain "high-risk" AI applications, but leaves significant gray zones for dual-use systems. Companies developing AI software for the consumer market are not required to declare whether the same technologies are being sold to government or military customers.

Europe and Italy: What Position in the Global Landscape of Military AI?

Europe finds itself in a paradoxical position. On one hand, the European Union has adopted one of the world's most stringent regulatory frameworks on artificial intelligence. On the other, NATO member states — including Italy — are under pressure to increase their military technological capabilities and reduce dependence on American or Chinese suppliers.

In Italy, the debate is still relatively under the radar compared to other countries. But there are significant developments:

• The National AI Plan 2026-2030 explicitly includes chapters on "national security" and "cyber defense," with a dedicated budget of over 400 million euros
• Leonardo S.p.A., the main Italian defense group, has launched partnerships with Italian AI startups to develop software systems for intelligence analysis and drone management
• The CISR (Interministerial Committee for the Security of the Republic) has commissioned studies on the integration of AI in national security systems

For Italian citizens and businesses, this means that the topic of military AI is not just an American or Chinese issue. It's a debate unfolding in Rome as well, though with less media fanfare.

The central question remains: can we develop artificial intelligence capabilities for national defense without sacrificing the ethical principles and fundamental rights that Europe has struggled to protect with the AI Act?

Frequently Asked Questions

Q: What exactly is "military AI" and how does it differ from civilian AI? A: Military AI uses the same algorithms and models as civilian AI but applied to defense, security, and intelligence contexts. It can include target recognition systems, predictive analysis, autonomous drone management, and surveillance. The main difference is not technological but concerns the application and ethical and legal implications.

Q: Can my smartphone data be used for military purposes? A: Directly, no — privacy laws (GDPR in Europe) protect users' personal data. However, the technologies developed thanks to user data (to train AI models) can later be adapted for military applications. It's a regulatory gray area that the European AI Act has not yet fully resolved.

Q: Which technology companies are most involved in military AI? A: The most exposed are Palantir (specialized in data analysis for governments), Microsoft (Pentagon contracts), Amazon Web Services (cloud infrastructure for defense), and Google (through separate divisions). In Europe, Leonardo, Thales, and Airbus Defence have developed internal AI capabilities.

Q: Does the European AI Act ban the use of artificial intelligence in military contexts? A: No. The AI Act focuses primarily on civilian applications. AI systems used exclusively for military and national defense purposes are explicitly excluded from the regulation's scope — a choice that has drawn criticism from many human rights organizations.

Q: What can citizens and consumers do to influence this debate? A: Consumers can educate themselves about the ethical policies of the technology companies whose services they use, support organizations advocating for stricter military AI regulation, and participate in public debate. As shareholders (even through pension funds or ETFs), you can vote for resolutions demanding greater transparency on military contracts.

Conclusion

Artificial intelligence "on the front lines" is not a metaphor: it is the reality of 2026, a reality in which the same software that powers your smartphone can, in adapted forms, guide weapons systems or fuel surveillance networks. Big Tech faces unprecedented pressure, caught between profit logic, government demands, and the ethical expectations of employees and users.

For those who follow the technology world, this is not a topic to delegate solely to geopolitics or military ethics experts. It concerns product choices, privacy policies, corporate governance, and ultimately, the kind of technological future we want to build. Italy and Europe still have the opportunity to play a leading role in this debate, but the time to do so is not unlimited.

Stay informed, ask uncomfortable questions of the companies whose products you use, and keep asking yourself: whose interests does this technology really serve?