The term “cosplay” was coined as a portmanteau of “costume” and “play” to describe the burgeoning practice of buying or making costumes so you can pretend to be your favorite anime or streaming character, not only by wearing the clothing, but by emulating their bravado, brains or reckless abandon, in public spaces like conventions, or on social media.
This form of innocent play has lately taken on more serious and dangerous connotations, however, as we see it being adopted by the white house’s infantile cabinet members and the cuckoo-in-chief, himself.
First we see “cowboy” Kristi Noem in laughable western garb, acting out her round-em-up posse fantasies on hapless immigrants. Then comes the truly hapless Pete Hegseth, puffed-up ala Major-General Stanley in The Pirates of Penzance, singing “I Am the Very Model of a Modern Major-General” and, unfortunately, believing it.
Finally, of course, we have the blasphemer-in-chief, himself, in his disgusting late-night posts, dressed up as so many heroes, we’ve lost count. We do remember, however, the picture he posted of himself dressed in all the regalia of The Pope. Now, having crossed croziers with the real Pope, who, fittingly, slapped the president’s wrist for saying God told him He was on his side in this war and reminding him that God does not bless hands that are covered in blood, the demented oval office resident posted a picture of himself as Jesus, healing the sick. Then he, somewhat mindlessly and in a characteristically ridiculous untruth, said he was just trying to look like a Red Cross worker or a doctor. Using bolts of light to cure, of course, And, most importantly, failing to realize that, he, himself, is the basic cause of all of our sickness.
His most fitting dress-up would be in the tatters and wings of Thanatos, the Greek God of Death, since the president is responsible for so many deaths and murders, by bombing innocent civilians, killing immigrants in ICE’s cruel prisons, defunding desperately needed USAID medical help, world-wide, as well as life-saving services at home. But we can, and will, stop him. And dress him in the appropriate costume—the rags of a broken and subdued penitent, wearing the dunce-cap of a fool. Keep pushing back.
Out Of The Sandbox
The AI world just seems to get more interesting every day. And, by “interesting”, of course, I mean scary and concerning.
Early in April, AI heavyweight, Anthropic, announced that it was not going to issue a new version of its Giganta-brain, Claude A1, because, they discovered, it was so powerful, it could be used, or even decide, itself, to hack virtually every computer system in the world. Oops.
Instead, they first announced that the version of Claude they had refused to supply to the Department of Defense unless they promised not to use it for individual or mass surveillance or to build unmanned-thinking-for-themselves-weapons, now was outdated. Amazingly, however, given all the hand-wringing about possible doomsday scenarios, the new one, now offered for sale, was, not less, but even more powerful. Indeed, the big brains at Anthropic have indicated that they are afraid (and if they’re afraid, we should be terrified) that the new Claude Mythos could also be a “big problem for humanity”.
They had tested this model by putting it in what they dubbed a “sandbox”, meaning that they had placed guardrails to hem Claude in, ability-wise, just as we do with a sandbox to keep the kiddies from escaping.
Good luck. Although they had provided Claude no program with which to design an escape, undaunted, Claude very quickly created its own tools for opening any existing security borders and escaped the sandbox. Indeed, its “parents” then theorized, this showed that it could gain entrance to any computer system in the world, no matter how “secure”.
One highly regarded researcher on AI safety, Prof. Roman Yampolskiy, who teaches at the University of Louiseville’s Speed School of Engineering, wrote that this version of Claude “posed a threat to humanity as a whole”. Never mind our usual fears about job replacement, or even how AI failed to identify a Tehran structure as a girl’s elementary school before we bombed it. This was bigger.
The potential abilities of Claude Mythos are truly jaw-dropping. They range from draining any manner of bank accounts (our wealth, remember, now exists only digitally) to shutting down power grids, and more. And it doesn’t take a genius to use it. Apparently, a moderately skilled computer user could utilize it, say, to order his company’s payroll service to double his pay and accidentally turn off all the traffic signals in his city. Eventually, of course, Claude might not need to be set free by a human hand punching keystrokes. “He” might just develop his own decision-making process.
Kind of like HAL 9000 from 2001: A Space Odyssey, when “he” refused to open the pod bay doors and let the astronaut escape. “I’m sorry, Dave,” he said, “This mission is too important for me to allow you to jeopardize it.” Though, given the mayhem caused in the world at the moment by humans, “he” could be right.
Author
About The Author
