Dell founder on "SHODAN" type perceptual AI: "It should remember the ozone layer.

General
Dell founder on "SHODAN" type perceptual AI: "It should remember the ozone layer.

We live in an age of AI hype. But while most of us are a little worried about what the rise of hyper-predictive text means for human creativity and critique, some in Silicon Valley are worried about Artificial General Intelligence (AGI). Or something like that.

But Michael Dell, founder and CEO of Dell, says there's no need to worry. In a recent virtual fireside chat with asset management firm Bernstein (discovered by The Register), Dell said he is "a little worried, but not too worried" about the advent of AGI. Why, because "as long as technology has existed, humans have worried about the bad things that could happen with technology, and about the terrible things that could happen ...... because we have told ourselves stories."

That worry, Dell continued, allows humanity to "create counter-actions" to prevent apocalyptic scenarios before they materialize. Remember the ozone layer. It didn't happen because humanity took countermeasures."

Dell (the man) continued that Dell's (the company) AI business is doing well. Customer demand has nearly doubled quarter-over-quarter, and the AI-optimized backlog has nearly doubled to $1.6 billion at the end of Q3," Dell (that guy again) boasted. As I write this as someone for whom "literal GLaDOS" is low on the priority list of fears, it sounds like something a tech CEO might say in the prologue to a movie about AI killing everyone.

Regardless, Dell doesn't think we need to worry about a robot uprising anytime soon. Except for climate change and nanoplastics in our blood. That and the fact that we didn't start repairing the ozone layer until there was a hole in it (2040 for repair, 2066 if you live in Antarctica). Maybe we're already doing this, but if you'll forgive me for a little commentary, it feels like reaching the right conclusion for the wrong reasons.

In my opinion, there is no need to worry about AGI. Because AGI is a spooky story dreamed up by wealthy technologists to hype up the capabilities of actual AI technology, and it is much cleaner and easier to deal with than the really scary thing about AI: the possibility that the entire creative industry will be destroyed and replaced by a homogeneous sludge of robots Because it's a story. Moreover, for all its problems, the Internet, despite being a genuinely useful storehouse of human knowledge, could become a large library of autocomplemented and totally incorrect nonsense, to the detriment of no one.

After all, I've already gotten to the point of adding "Reddit" to most of my Google searches to make sure I'm actually getting human input on whatever issue I'm facing. And it's a far more vexing problem with a profit-threatening solution than a HAL9000 nuisance.

Categories