Astrophysicists Michael Hippke and John G. Learned considered this possibility in a paper published last year to the preprint server arXiv. Space is big, high-speed travel is difficult, and fleets of battleships are expensive, they reasoned, so the likeliest mode of attack for hypothetically malicious aliens would be code concealed in a message. Such code could contain an advanced A.I. that could sneak into computer systems and spread throughout the Internet, or a virus that would destabilize banking systems and electrical grids. Or perhaps, they suggested, an alien attack could simply be a panic-inducing prank, a statement like “We will make your sun go supernova tomorrow.”

The only way to avoid this outcome is not to open a message from extraterrestrials, the cosmic equivalent of deleting a suspicious email.

“A message… cannot be decontaminated with certainty,” Hippke and Learned cautioned. “For anything more complex than easily printable images or plain text, the technical risks are impossible to assess beforehand. We may only choose to destroy such a message, or take the risk. The risk for humanity may be small, but not zero.”