On Sept., 26, 1983, shortly after midnight, the Soviet Oko nuclear early warning system detected five missiles launched from the United States and headed toward Moscow. Stanislav Petrov, a young lieutenant colonel in the Soviet Air Defense Force, was the duty in the Serpukhov-15 bunker that housed the Oko command center. Petrov was the man in charge of alerting the Soviets about a nuclear attack, which would trigger a retaliatory strike.
He determined the Oko had likely malfunctioned and the alarm was false. The Americans would not start World War III with a quintet of missiles (risking total annihilation). It was a daring judgment call. He was, of course, right. As the U.S. prepares to undertake a new nuclear posture review to determine the future direction of the nation’s nuclear weapons, a report from a United Nations research institute warns the risks of a catastrophic error—like the one that took place that early morning in 1983—are growing, not shrinking. Next time, there may be no Lt. Col. Petrov in place to avoid a catastrophe.
On Monday, the Defense Department commenced a new, massive study into its nuclear weapons arsenal, looking at how weapons are kept, how the U.S. would use them in war and whether they present an intimidating enough threat to other countries not to attack us. The review was mandated by President Donald Trump in a Jan 27, memo.
The Pentagon is scheduled to complete the review by the end of the year, an essential step as the military seeks to modernize different aspects of its nuclear deterrent. But a new report from the United Nations Institute for Disarmament Research, or UNIDR, argues as the modern battlefield becomes more technologically complex, crowded with more sensors, satellites, drones and interconnected networks, the risks of another nuclear accident are on the rise.
“A greater reliance on automated systems can lead to misplaced confidence while introducing new points of vulnerability,” says the report. Those new points of vulnerability include so-called hidden interactions. That means a sensor or computer program misinterpreting some bit of data and possibly presenting false information in a way that could cause an accident. The 1987 incident provides a good case in point. Oko satellites mistook a very unusual sunspot on top of a high-altitude cloud as a missile strike, hence the false alarm.
Take those satellites, combine them with sensors on drones and data from other sources as well, including new, perhaps unproven technologies to detect missile launches and the picture becomes much more crowded and murky.
“The complex interactions and tightly coupled systems linked to nuclear arsenals (like those for early warning, and launch command and control) have made ‘accidental war more likely’” the report’s authors say.
Add to that the fact the number of states that have access to nuclear weapons is increasing, and the number of platforms they might be able to use to deliver those weapons is also going up. Consider the controversial U.S. plans for a long-range standoff weapon, or LRSO, basically a big nuclear cruise missile that can be fired off a fighter jet. Reports have surfaced that the U.S. is even considering nuclear-armed drones (that would be remotely operated by human pilots and the degree of seriousness in the considerations is up to debate).
Those might sound like awesome capabilities, but they increase the chances of a nuclear accident or retaliatory strike, according to the authors of the report, because such weapons essentially turn every jet and drone into a potential nuclear threat in the eyes of an adversary.
“The spread of other systems, such as cruise missiles and drones, and their increasingly frequent use in military conflicts can also add to the complexity of the situation, as can the development of capabilities to detect missiles,” writes Russian arms expert Pavel Podvig in the report.
Rising tensions and eroding trust between the U.S. and Russia don’t bode well stability.
“The intensity of interactions between the United States and Russian militaries, although lower than during the height of the Cold War, does not show signs of decreasing,” he notes.
The digital interlinks that will emerge between nuclear weapons and other systems are another threat factor.
“The myth that nuclear facilities and platforms are air-gapped—meaning they are not connected to the Internet—is gradually decreasing as nuclear establishments are increasingly informed about cyber threats,” write information security researchers Patricia Lewis and Beyza Unal in the report.
They identify several potential cyber vulnerabilities in nuclear command and control systems—not in nuclear weapons themselves—but in the technologies around them and the methods operators use to control those weapons or make decisions about using them. The key targets the researchers identify include: communications between command-and-control centers, targeting information from ground stations, and “robotic autonomous systems within the strategic infrastructure.”
The U.S. and other nations may have closed those portholes, or partially closed them, or not at all. The lack of transparency about the nuclear exercise makes it impossible to know and increases risk.
“Academic and policy analysis on nuclear weapons facilities and nuclear command and control systems is still in its infancy, due to the secrecy surrounding nuclear weapons systems and the underinvestment in research into potential problems until recently,” they write.
But again, those are just potential vulnerabilities. The U.S. is currently modernizing the communications network it would use in the event of a nuclear conflict, the Minimum Essential Emergency Communications Network, or MEECN, to improve efficiency. That sounds great, however, every new system, center, or device connected to the MEECN is a potential vulnerability.
Today, it’s impossible for an adversary to hack into the current the ICBM arsenal. But that may not be the case forever. Military researchers anticipate that future ICBMs, as well as bombers, will exhibit “some level of connectivity to the rest of the war-fighting system,” Werner J.A. Dahm, the chair of the Air Force Scientific Advisory Board told Defense One last December.
Importantly, an adversary does not even have to exploit a cyber vulnerability either in nuclear weapons or the communications lines or data gathering sensors around them in order to create a cyber nuclear crisis. They just have to convince the adversary they have that ability.
“Loss of trust in technology has further implications for attribution and strategic calculus in crisis decision-making and may increase the risk of misperception," write Lewis and Unal.