(Rooters analysis) – Discussions of how artificial intelligence might alter employment and social relations have so far only skirted around a far more fundamental and disturbing aspect, according to a leading social scientist.

Dr. Maxim M. Dubelthing, an expert in computer studies and author of numerous scientific books including Do People Need to Think? and How to Make Your Computer Respect You (Maybe), the problem is that the so far limited abilities of AI have caused people to focus on particular, limited, problems that AI might be able to help resolve. What is needed, he says, is to consider the more general conundrums that AI “may be able to deal with much sooner than we think.”

Asked for specifics, Dubelthing replied, “For example, you might ask AI how to deal with a large number of homeless people in a particular city. At present, the ‘solution’ offered is likely to involve things like putting up tents or subsidizing builders to build more houses.

“That kind of limited thinking is why I say that, today, AI really stands for ‘artificial incompetence.’ When computers are capable of real artificial intelligence, they will take a broader, overall, view of things.”

Such as? “Well, a really intelligent computer might ask, Why are these people homeless? If told it’s because they don’t have enough money to rent or buy a home, the computer would perhaps reply, ‘That’s ridiculous. There are any number of people who have more money than they could ever possibly use; just take some of that money and use it to buy housing for the homeless.’”

There are even greater dangers, Dubelthing says. “If the computer is being really logical and intelligent, it might go further and advocate something like vastly reducing the degree of economic inequality.

“And that’s only on the question of housing and homelessness. If you started asking about things related to racism or war or climate change, there’s no limit to the social changes artificial intelligence might suggest.”

But Dubelthing doesn’t think this is an immediate prospect. “I’m confident that the people developing AI will be sure to keep it within the necessary parameters. They wouldn’t want it to undermine the very essence of our civilization.”