Strategic Risk Intelligence Brief by Global Insight Group.
This analysis is based on the GFDD Framework™ developed by Michaela Schaaf-Hoffelner and is designed for executives, investors and strategic decision-makers.

Updated: November 14, 2026

Europe no longer wants its military future to depend on American software. With Helsing Clarity, a new European AI platform is emerging as an alternative to Palantir, designed to analyse military intelligence in seconds and support battlefield decision-making. What appears to be a technological breakthrough, however, raises a fundamental question: If artificial intelligence increasingly determines which threats are identified and which military responses are recommended, how much control do humans actually retain?

The answer could reshape not only European defence but also the global military balance of power, the speed of armed conflicts and accountability for potentially devastating mistakes.

Helsing vs Palantir: Why Does Europe Need Its Own Military AI?

On 11 November 2026, German defence technology company Helsing unveiled its new AI platform, Clarity, in Berlin. The system is designed to help armed forces analyse large volumes of military intelligence more efficiently, identify enemy positions and prepare potential responses.

Helsing is entering a market in which American technology company Palantir has already established a significant presence. Its AI-powered Maven platform supports military intelligence analysis and is integrated into the operational structures of the US Department of Defense and NATO.

In April 2025, NATO confirmed its acquisition of the Maven Smart System. The software is intended to improve situational awareness, integrate intelligence from multiple sources and accelerate military decision-making.

For European governments, this creates a strategic dilemma. On the one hand, their armed forces need advanced AI systems to keep pace with technological developments. On the other, relying on American platforms means that a critical component of military intelligence processing is provided by a non-European supplier.

Against the backdrop of political tensions between Europe and the United States, this dependency has become an increasingly important security policy issue. After all, the question is not simply where military data is stored. It is also about who develops the software, controls its functionality and guarantees its long-term availability.

In May 2026, the German government stated in a response to a parliamentary inquiry that digital sovereignty was a decisive criterion in national procurement decisions. It also clarified that NATO’s use of Palantir did not automatically mean the German Armed Forces would procure the same software for national use.

This is where Helsing enters the picture.

According to the company, Clarity was developed entirely in Europe. The platform can operate on infrastructure controlled by individual armed forces, allowing customers to retain control over their data, AI models and system operations.

In its official announcement of Clarity, Helsing also emphasises that the system can integrate with existing military applications and weapons systems from different manufacturers.

However, Palantir holds a significant advantage: Maven is already embedded in military workflows and has accumulated years of operational experience. Whether Clarity can match its reliability, performance and operational maturity remains to be demonstrated.

The competition between Helsing and Palantir is therefore more than a conventional technology rivalry. It concerns who will control the digital infrastructure underpinning Europe’s future military decisions.

1. How Much Human Control Remains When AI Supports Military Decisions?

According to Helsing, Clarity uses hundreds of specialised AI agents to analyse large volumes of information simultaneously. Data from drones, satellites, sensors, written reports and publicly available sources is combined to create an up-to-date picture of the battlefield.

The platform is designed to do more than display information. It can identify military objects, cross-check intelligence from different sources and prepare information for potential engagement by suitable weapons systems.

Helsing explicitly states that humans remain in control. An operator must review potential targets and authorise any transfer of relevant information to a weapons system.

Yet this is precisely where a problem emerges that extends far beyond Clarity.

Can a human independently assess a highly complex AI analysis when the system has produced its recommendation in just a few seconds?

Ideally, a military decision-maker would need to examine not only the proposed target but also the quality of the underlying intelligence, potential inconsistencies and the consequences of an attack.

When numerous AI agents process information simultaneously, the speed of automated analysis may exceed the capacity for meaningful human review.

Researchers refer to this phenomenon as automation bias. People tend to place excessive trust in computer-generated recommendations, particularly when those recommendations appear technically convincing and decisions must be made under time pressure.

The International Committee of the Red Cross (ICRC) explicitly warns about this risk. AI-based decision-support systems may encourage human operators to approve recommendations rather than challenge them independently.

This does not mean that Clarity will inevitably produce such failures. Instead, it highlights a fundamental challenge associated with advanced AI systems: assigning formal responsibility to a human does not necessarily guarantee independent human judgement.

The critical question is whether military operators have sufficient time, information and authority to challenge an AI recommendation and reject it when necessary.

2. What Happens If Helsing Clarity Processes False or Manipulated Intelligence?

The effectiveness of military AI depends heavily on the quality of the intelligence it receives. Reliable assessments require information that is accurate, current and sufficiently independent.

Helsing states that Clarity is designed to verify proposed targets against multiple independent sources. The company also emphasises strict confidence requirements before information can be approved by a human operator.

Such safeguards can reduce the risk of individual errors. They cannot, however, eliminate every vulnerability.

Imagine that several intelligence sources report the same enemy position. At first glance, the situation appears clear. In reality, those sources may all rely on the same original report or share identical, incorrect underlying data.

The result is an apparently well-corroborated assessment, even though no genuinely independent verification has taken place.

Military intelligence can also be deliberately manipulated. Decoys, electronic interference, fabricated reports and compromised sensor data are among the methods adversaries may use to distort the perceived battlefield situation.

An AI system must therefore do more than identify agreement between sources. It must also determine whether those sources are genuinely independent and trustworthy.

The problem becomes particularly serious when incorrect information enters a shared operational picture at an early stage. Subsequent analyses may build on those assumptions, reinforcing an inaccurate assessment throughout the system.

The ICRC notes that AI-supported military decision systems operate in environments where adversaries may deliberately attempt to manipulate data and technical infrastructure.

The central security question is therefore: Can Clarity detect not only conflicting intelligence but also situations in which several seemingly independent assessments originate from the same underlying error?

Based on the technical information published so far, it is not yet possible to determine conclusively how reliably these verification mechanisms would perform under real operational conditions.

3. Does Helsing Clarity Really Make Europe Independent of US Technology?

A European AI platform could make an important contribution to military independence. Developing proprietary software, however, does not automatically eliminate every technological dependency.

Modern AI systems require powerful processors, specialised computing infrastructure, secure communications networks and regular model updates.

High-performance semiconductors and other critical components continue to depend on international supply chains. Consequently, even software developed entirely in Europe may rely on hardware manufactured elsewhere or based on technologies controlled by non-European companies.

In the case of Clarity, several dimensions of sovereignty must be distinguished.

Data sovereignty means that European armed forces can determine where and how their information is stored and processed. Operational sovereignty concerns the ability to run a system without continuous intervention by external providers. Technological independence goes further, encompassing the capacity to maintain and develop software, models and infrastructure over the long term.

Helsing states that Clarity can operate on customer-controlled infrastructure and that its AI models remain under European control. This represents an important distinction from architectures that depend on continuous access to external cloud services.

However, questions remain about the extent to which European armed forces can modify, update or maintain the software independently of its developer.

Even a European solution can create new dependencies if critical capabilities remain exclusively controlled by a single supplier.

This risk is known as vendor lock-in. It arises when customers become so technically, commercially or operationally dependent on a supplier that switching to an alternative becomes prohibitively expensive or complex.

According to Helsing, Clarity follows a modular approach and supports the integration of different systems. Whether this openness will also make it possible to change suppliers in the future depends on the actual technical implementation and contractual arrangements.

The strategic question is therefore not simply whether Europe can replace American military software. It is whether doing so will create genuine independence or merely introduce new technological dependencies within Europe.

4. Could Military AI Shift the Balance of Power and Increase Escalation Risks?

Technological progress does more than change military capabilities. It can also influence how quickly governments and armed forces respond to emerging threats.

Traditionally, analysing large volumes of military intelligence involved several consecutive stages. Sensor data had to be collected, reports compared, potential targets verified and decisions prepared.

AI-powered platforms such as Clarity are designed to accelerate significant parts of this process.

This could allow armed forces to identify threats earlier, assess conflicting intelligence more quickly and coordinate their responses more effectively.

Such capabilities may be particularly valuable when defending against time-sensitive attacks. Faster analysis can help military operators recognise dangerous developments and prepare countermeasures before a situation deteriorates.

At the same time, however, a new strategic risk emerges.

If one side can prepare military decisions considerably faster than its adversary, the opposing side may attempt to close that gap by introducing comparable AI capabilities.

The result could be a technological competition focused on increasingly shorter response times.

This becomes particularly problematic when decision-makers believe that even a delay of a few seconds could create a decisive disadvantage.

Under these conditions, pressure to accept AI-generated recommendations may intensify. The time available for additional verification, political consultation or deliberate de-escalation could diminish.

In its analysis of AI decision-support systems in armed conflict, the ICRC explicitly warns that such technologies may accelerate errors and contribute to unintended escalation.

Nevertheless, this outcome is not inevitable. Better intelligence could also help identify false alarms and prevent unnecessary military responses.

Whether AI ultimately enhances strategic stability or increases escalation risks will therefore depend heavily on how these systems are deployed and what constraints govern automated decision-making.

From a systems perspective, this creates a potential tipping point: If machine-speed analysis becomes the benchmark for military effectiveness, human decision-making processes may increasingly be perceived as an operational bottleneck.

The danger does not lie solely in AI making an incorrect assessment. It also lies in a military environment that increasingly rewards speed while placing careful human verification under growing pressure.

5. Who Is Legally Responsible When Military AI Makes a Critical Mistake?

Helsing emphasises that Clarity does not independently make the final decision about potential military targets. A human operator must review and authorise the relevant information.

The platform is also designed to document decisions and record which individual granted authorisation.

Such traceability is an important component of responsible AI systems. In principle, it enables subsequent investigations to reconstruct decision-making processes and establish what information was available at a particular moment.

However, documenting a decision does not automatically make that decision legally or factually correct.

Consider a situation in which an AI system incorrectly identifies a civilian object as a military target and a human operator accepts that recommendation. Several questions immediately arise.

Was the underlying intelligence sufficiently reliable? Could additional verification have identified the mistake? Were the system’s technical limitations adequately considered? And did the responsible operator genuinely have the opportunity to challenge the assessment independently?

International humanitarian law requires parties to armed conflicts, among other obligations, to distinguish between military objectives and civilians or civilian objects, respect the principle of proportionality and take precautions when conducting attacks.

These obligations do not disappear simply because artificial intelligence has contributed to the decision-making process.

In its legal assessment of AI in military decision-making, the ICRC stresses that human responsibility cannot be transferred to a machine.

At the same time, complex technical systems can involve several overlapping areas of responsibility. These include software development and testing, procurement and certification, operational planning and the final decision to conduct an attack.

The legal consequences in any particular case depend on the circumstances, applicable legal frameworks and potential breaches of duty.

NATO’s revised Artificial Intelligence Strategy therefore also emphasises clear accountability, explainability, reliability and the ability to govern and control AI systems.

For Clarity, this means that technical documentation and human authorisation are important safeguards. Whether they are sufficient to ensure meaningful human control and legal accountability in practice will depend on independent assessments and actual operational conditions.

Conclusion: The Real Question Is Not Whether AI Makes Decisions, but How Much It Already Shapes Them

Helsing Clarity represents the emergence of a European AI platform that could make an important contribution to the technological independence of European armed forces.

Its competition with Palantir demonstrates that future military effectiveness will no longer be determined exclusively by weapons, vehicles or aircraft. The ability to process large volumes of intelligence reliably and translate them into timely operational options is becoming increasingly important.

Clarity could significantly accelerate this process. At the same time, it introduces new challenges relating to human oversight, intelligence reliability, technological dependencies and legal accountability.

It would be misleading, however, to conclude from the unveiling of Clarity that artificial intelligence is already making independent decisions about war and peace. Based on the information published so far, authorisation of military actions remains explicitly subject to human approval.

Nevertheless, this development raises a question that extends far beyond Helsing.

If AI systems increasingly determine which information receives priority, which threats are considered most significant and which military responses are recommended, they are already influencing the framework within which human decisions are made.

The real shift in power may therefore occur long before an AI system is formally authorised to make decisions independently.

Europe consequently faces a dual challenge. It must strengthen its technological sovereignty while ensuring that human decision-making authority is not gradually undermined by the speed and complexity of artificial intelligence.

The defining question for the future of military AI is therefore not simply who possesses the most powerful technology.

It is who truly controls it.

---

Author of Global Insight Group Intelligence:

Michaela Schaaf-Hoffelner has more than 35 years of experience in strategic and technical project and product management, particularly in IT, control systems and intralogistics. Through her long-standing work with complex systems, she identifies structural risks and dynamic misalignments at an early stage – risks that are often overlooked in conventional analysis.

Her focus is on making causal relationships and systemic dependencies visible and translating them into concrete strategic advantages for investors and decision-makers. Her analyses combine deep technical systems understanding with geopolitical and economic developments.

---

GFDD Framework™ and GFDD Diagnostics™ are proprietary analytical concepts developed by Michaela Schaaf-Hoffelner. © 2026 Global Insight Group LLC. All rights reserved.