The recent violations of Allied airspace represent an unprecedented escalation of tensions between NATO and the Russian Federation. Neither the number nor the type of incursions has been seen before, and it has led to a strong and determined response by the Alliance. NATO must try to protect its airspace against unauthorised activities by foreign military aircraft or unmanned aerial vehicles. Yet multiple factors make this task difficult.
First, NATO members face different levels of exposure, and geography plays a crucial role. Second, past and present reactions to violations often lack consistency, which diminishes the effectiveness of future deterrence. Finally, the use of low-cost technologies in (hybrid) warfare poses significant challenges for modern societies reliant on complex networks of critical infrastructure.
There is no easy short-term solution. NATO and its member states must boost air defence capabilities and strengthen the protection of infrastructure, but this will not alter the opponent’s calculus. Alongside deterrence by denial, the Alliance must also employ deterrence by punishment.
NATO cannot dissuade further violations without imposing costs on Russia after every incursion. The magnitude of punishment has to depend on the type of incident, but it cannot be skipped. The Alliance must differentiate between hostile incidents and those resulting from mere ignorance, between various locations and critical infrastructure sites, as well as between drones and aircraft. However, it still needs to react rapidly, repeatedly, and consistently in similar situations over time and across member states, if it wants deterrence to be effective.
Geography makes a difference
When up to twenty Russian drones violated Polish airspace in early September, most ended up in borderlands near Ukraine, and a few were shot down. Others flew deep into Poland before crashing in regions that host several important Air Force bases, including Malbork, Mińsk Mazowiecki, and Łask. Given the severity of the incident, Prime Minister Donald Tusk´s warning that future entries of any unauthorised objects would be met by force seems entirely justified. Shooting down an intruding Russian aircraft requires some willingness to escalate, but the warning was communicated. To paraphrase Polish Foreign Minister Radosław Sikorski, an eventual ‘whining’ of Moscow about a lost aircraft at the UN Security Council is avoidable.
Poland is in a similar situation to Turkey in November 2015, when it shot down a Russian Su-24. Then, the Russian Air Force was conducting combat missions over neighbouring Syria in support of Assad. Today, Poland borders Ukraine, which has been facing Russian aerial attacks for nearly four years. Due to this proximity to an active combat zone, any Russian plane violating Polish airspace implies a significantly greater risk to national security. Thus, it is a legitimate target from a military perspective.
Geography, as one of the key factors determining an appropriate response, varies among NATO members. Estonian airspace was violated days after the incident in Poland, but although Estonia is one of the most exposed countries on NATO’s Eastern Flank, it does not border Ukraine. Associated risks are, therefore, smaller, allowing for a more nuanced approach under lower time pressure. A default assumption about Russian military aircraft violating Estonian airspace is that they are not on a combat mission. Hence, the situation was markedly different compared to the Polish case.
Moreover, flights between mainland Russia and Kaliningrad must pass through the Gulf of Finland via a narrow corridor of international airspace, divided into three Flight Information Regions, which are managed from Helsinki, Tallinn, and Saint Petersburg. It is a historical anomaly that airspace over Estonia’s tiny uninhabited island of Vaindloo in the Gulf of Finland is administered by Petersburg, while Tallinn manages an adjacent patch of airspace over Russian waters.
The shortest route from Russia to Kaliningrad passes directly over Vaindloo. The International Civil Aviation Organisation rules require state-owned aircraft to obtain permission before entering foreign airspace; however, Russian military aircraft routinely disregard this requirement. Instead of detouring, they often cut across Estonian airspace for a few seconds without radio contact with Tallinn. The September incursion by three MiG-31s followed the same pattern.
Bad signalling
NATO’s response to airspace incidents significantly impacts its ability to deter future incidents. In response to the incident in Polish airspace, Warsaw requested consultations under Article 4 of the Washington Treaty. This led to the launch of Operation Eastern Sentry, reinforcing air defence in Poland and on the Eastern Flank with additional air and naval assets. It was an entirely justified, even if extraordinary, step. Article 4 has been invoked only seven times in NATO’s history, including after Crimea’s annexation and the 2022 invasion of Ukraine.
Such strong signalling should remain exceptional, as overuse weakens its impact. Thus, when Estonia invoked Article 4 just a week after Poland, many NATO capitals were surprised. Since joining NATO in 2004, Estonia has suffered more than sixty Russian violations, about 80% over Vaindloo. Until September, Tallinn responded by publicising the incidents and summoning Russian diplomats, never by invoking Article 4.
The September incursion was unusually severe, as three aircraft were in Estonian airspace for twelve minutes. Yet Estonia had faced serious violations before, such as the 2021 incursion by two Su-35s during NATO’s BALTOPS exercise, without invoking Article 4. The incursion was also unusually long because MiG-31s followed the Vaindloo-Naissaar-Osmussaar line at the edge of Estonian airspace. The airspace close to Naissaar and Osmussaar islands was also violated in November 2004 and June 2007, but Tallinn did not invoke Article 4 at that time either.
In the latest incursion, Italian F-35s conducting Baltic Air Policing from Ämari Air Base identified the intruders and escorted them out of Estonian airspace. After assessing, among other things, the flight path and armament of MiG-31s, they determined that the intruders did not represent an imminent danger. Why Tallinn decided to invoke Article 4 at that time remains a puzzle. Understandably, it wants to stop Russia’s routine violations, but invoking Article 4 under these circumstances risks diluting its future value.
The inconsistency of approach within NATO becomes even sharper in light of Norway’s post-hoc disclosure that Russia violated its airspace high in the north already three times in 2025, without ever triggering Article 4. Such uneven signalling makes deterrence harder.
Appropriate response matters
Geography and past practices form the backdrop, but technical and operational choices ultimately shape the effectiveness of deterrence. The response to the violation of Estonian airspace on 19th September was a textbook alpha-scramble (an order for fighter jets to take off immediately due to suspicious activity in the air). But this is not typical. Only a small fraction of Estonian airspace violations trigger alpha-scramble orders from NATO’s Combined Air Operations Centre in Uedem. Allied quick reaction alert assets intercept intruders with or without a prefiled flight plan, with or without a squawking transponder, and with or without radio contact. This lack of consistency erodes deterrence.
Drone incursions are even more challenging when it comes to adopting an appropriate response. Russian drones over Poland and unidentified UAVs near Danish airports and military sites illustrate the problem. Condemnations from NATO have little effect in Moscow, and the use of fighter jets, advanced air-defence systems, and AWACS against cheap Russian drones may be effective, but it is certainly not efficient. Gerbera drones cost about $10,000 each and are often used as decoys, overwhelming Ukrainian defences. By contrast, operating a fifth-generation fighter or firing an air-to-air or surface-to-air missile costs exponentially more.
NATO currently lacks efficient means of countering low-cost drones. The dilemma is well known, and the solution requires a combination of electronic-warfare capabilities, interceptor drones, and the ability to shoot down low-flying objects for less than their production cost. Some promising systems exist, but comprehensive solutions will likely take years to be fielded. The Danish case illustrates that complete protection of critical infrastructure is unrealistic. Even the latest Eastern Flank anti-drone initiative would be better described as a loose net than a wall. Modern societies rely on a dense network of critical infrastructure sites, making it nearly impossible for an effective and efficient air defence system to provide full coverage.
What we tend to forget is that effective defence is only one part of the deterrence equation. Deterrence requires both denial and punishment. It is not a single-shot game, but a repetitive struggle, especially below the conventional level. Without the imposition of costs, adversaries will continue probing. Reliance on defence without (sub-)conventional punishment will prevent a deterrent effect from accumulating. Incidents will continue to occur due to the low stakes for the adversary and the numerous vulnerabilities in target countries.
Many NATO member states, however, are apparently unwilling to turn the tables and match the Russian escalation on equal terms. Deterrence requires certainty and swiftness of punishment, while severity is of secondary importance. It is not about shooting at every single drone and military aircraft, but elementary consistency in responding to aerial incidents might help. Different incursions must be treated differently, but there needs to be consistency both over time and across member states. Inconsistency is a bad signal that decreases credibility and invites further testing.
The magnitude of punishment can differ depending on the type of incident. Yet the imposition of costs not only on individual intruders but also on relevant group actors must be a given. Intelligence flights around Kaliningrad Oblast after every incursion would be a good starting point, and the Russian military has already demonstrated that it does not appreciate this kind of information gathering. But one of the best options would be to redirect a few low-cost surveillance drones toward Kaliningrad itself. Even if none return, they would deliver a more effective (and efficient) deterrent message than expending expensive AIM-120 missiles on cheap Russian drones. This would also provide useful information on Russian air defences.
The European Leadership Network itself as an institution holds no formal policy positions. The opinions articulated above represent the views of the authors rather than the European Leadership Network or its members. The ELN aims to encourage debates that will help develop Europe’s capacity to address the pressing foreign, defence, and security policy challenges of our time, to further its charitable purposes.
Image credit: Pacific Press Media Production Corp./Alamy Live News
