Why AI collaboration is necessary.
Artificial intelligence has advanced to a stage where its potential to disrupt or uplift societies rivals that of nuclear power. Yet as global powers race to develop AI-driven weaponry and sophisticated autonomous systemsI can’t help but think: if AI becomes a tool of war, it will be as unwinnable and destructive as nuclear conflict. In a traditional war, overwhelming force can produce a decisive victory, but once AI has spread across multiple actors, it is no longer a tool controlled by one power alone. Instead, it becomes a universal capability that any sufficiently advanced individual, nation or organization can harness. In such a scenario, each player’s capacity for destruction mirrors that of its adversaries, creating the same precarious standoff once exclusively associated with nuclear arsenals.
Nuclear weapons introduced the concept of mutually assured destruction, suggesting that once two or more adversaries possessed these weapons, any nuclear strike would invite catastrophic retaliation. No one dared to press the button first because the ensuing fallout would obliterate all sides.
AI, especially with the lightning-fast improvements in machine learning algorithms and automated systems, may prove even more unstable. Where nuclear weapons require cumbersome stockpiles, specific delivery mechanisms, and relatively transparent testing procedures, AI development can happen behind lines of computer code through open-source communities. Because the barriers to entry are significantly lower than those for nuclear weapons, more actors can gain access to extremely powerful AI systems capable of crippling infrastructure, launching disinformation campaigns, or even commandeering automated military hardware. This level of potential chaos invites a stalemate dynamic that could paralyze any aggressor, for launching the first strike might guarantee severe retaliation by equally capable adversaries.
The nature of AI-driven warfare only heightens the potential risk of miscalculation. In the nuclear era, satellites and international monitoring regimes offered at least some level of transparency about who possessed what arsenal and where. AI capabilities, in contrast, can be hidden behind encrypted servers, masked by innocuous commercial applications, or even piggybacked on consumer technologies. This lack of transparency fosters suspicion and escalation. Just as nations felt compelled to build extensive nuclear arsenals out of fear of being outgunned, countries now feel pressured to push AI research to stay ahead of potential rivals. But the more advanced AI becomes, the broader its capacity for both offensive and defensive strategies. Every nation that deploys AI for security or power projection inadvertently pushes competitors to respond in kind, creating a cycle that is neither easily monitored nor regulated.
An AI arms race spiraling out of control carries the dire possibility of unintended accidents as well. Malfunctioning or misaligned systems could make decisions at a scale and speed beyond human comprehension. Military AI might inadvertently identify false targets or launch misguided cyberattacks without the deliberate oversight of human command. In nuclear standoffs, at least a series of human checkpoints stands between the first sign of incoming missiles and the final launch codes. AI could accelerate this decision-making window to mere seconds, leaving little to no time for human introspection or diplomatic resolution. In a conflict setting, the risk is that a single glitch, misinterpretation, or overzealous algorithm might trigger a rapid escalation that human operators can’t contain in time.
Given these dangers, the question emerges: who would be the first to initiate an AI-based attack when doing so might unleash uncontrollable chaos? Rational actors, as in the nuclear era, may realize that no viable path to victory exists if both sides are similarly equipped. Under such circumstances, a stalemate forms where neither party can gain from aggression, effectively deterring conflict—or at least forcing it into covert channels rather than overt warfare. Many leaders hope that this deterrent effect will hold, much as it has for nuclear weapons. Yet the clandestine nature of AI research leaves far more room for miscalculation, sabotage, or ideological zeal that defies the logic of deterrence.
It won’t be easy but a new frameworks must evolve to govern AI’s development. Failing to do so means walking a dangerous line where a future conflict could quickly escalate beyond human control, leaving nothing but mutual destruction in its wake.
Suggesting collaboration apparently makes me a communist.These idiotic labels, I am a survivalist and possibly the devil whispering in your ear. I also have kids and am deeply concerned about THEIR future.
Thanks for reading and sign up to our newsletter.
2ndrevolution.org