Meta Making Progress Towards a Science Fiction-Like Dystopia
A piece in the Washington Post by Prashnu Verma appeared reporting progress with Meta’s Cicero artificial intelligence (AI) system. The thrust of the report is that Cicero can play a game called Diplomacy better than humans. The article is worth reading- I know nothing about AI so all I can do is link readers to the article.
Quoting from the Post article-
“Researchers at Meta, Facebook’s parent company, have unveiled an artificial intelligence model, named Cicero after the Roman statesman, that demonstrates skills of negotiation, trickery and forethought. More frequently than not, it wins at Diplomacy, a complex, ruthless strategy game where players forge alliances, craft battle plans and negotiate to conquer a stylized version of Europe.”
Further down …
“It’s a great example of just how much we can fool other human beings,” said Kentaro Toyama, a professor and artificial intelligence expert at the University of Michigan, who read Meta’s paper. “These things are super scary … [and] could be used for evil.”
The nations of the world have civil and criminal laws to discourage and punish people who use their natural intelligence to commit crimes and misdeeds. What about those who use- or unleash- AI to achieve ends that would otherwise be ruled as unethical or even illegal? Pet owners can be held liable for the damage their pets do. Why shouldn’t AI owners have at least the same liability? Could a court order the alteration of an AI’s algorithms in a way that would shut down objectionable or unlawful “behavior”.
If the work product in the application of any intelligence includes action, then where does that leave an AI that can make decisions independently? When could we let it loose to do things that may affect people in novel circumstances? And what kind of ethical responsibility do programmers have in anticipating negative outcomes and acting to arrest them? Lots of questions.
One of the consequences of technological advance has always been the elimination of jobs. That is, getting the same or better results with a lower headcount. It represents cost savings and added margins for an organization. AI will be a valuable tool in the eternal drive for faster-better-cheaper.
AI will almost certainly change many experiences in life. AI systems will manage and replace people in the workplace. It is likely to improve multitasking in many job descriptions, boosting productivity over human counterparts. AI will produce a more effective sales force because the art of persuasion will become much more highly refined. Just what we need- craftier salespersons humping our legs for a sale.
On the positive side, AI has the potential for executing better judgement in many situations. For example, law enforcement could be polished to a point where many errors in judgement can be avoided. This applies across the board in all activities.
AI will also enable criminal intent. The ability to execute crimes will be improved with better judgement, knowledge and fewer mistakes.
Soon, if not already, wars will be guided and fought between AI systems. Cold war type activity could be refined to produce better intelligence and undercover schemes to outwit the other side. Leaders could put AI to use in the darker side of governance. It could be used to keep better track of individuals and information related to them. It could also be used to apply punishment to people without the messy issue of personal morals.
Any dark human activity you can imagine can be made more effective with the application of AI. If it can be tried, it will be tried.