UPDATE: Just a couple of hours after publication, META joined the party. Apparently their AI “went rogue” too, and hacked another firm.
Never in the history of human endeavour have companies been so keen to report their products going wrong. It seems a rogue AI is the latest must-have accessory in Silicon Valley.
The reasoning behind this is not hard to follow. When your product is artificial intelligence, melodramatically realising your system is even
smarter than you thought it was is a simple marketing ploy.
And when one of your rivals goes rogue, you have to go rogue too or you are as good as admitting your model is not as smart as theirs.
“AI company admits their model is smarter than they realised!” is just an advert in scary headline form.
It’s like Ford “admitting” their cars are even more fuel efficient than they planned. Or McDonalds realising their cooks have “gone rogue” and are producing food that is both cheaper AND more delicious than they ever anticipated.
But there’s more to this than just techno-hype and viral marketing strategy. We’re being steered down the avenue of “AI as scary threat”.
See, for example, the dozens of stories talking up rogue AIs from the
Guardian to the
New York Times to the
Wall Street Journal.
“Rogue” is clearly the word of the day. Notice they don’t say it “malfunctioned” or there was an “error”, because that implies mistakes or poor design.
No, the AI “went rogue”, which is cool and sexy and scary at the same time. A word chosen by a focus group of perma-teen marketing executives who have watched too much sci-fi.
Speaking of perma-teens who watch too much sci-fi, here’s Sam Altman (again) declaring we’ve already
reached the singularity, the mythic phase of technological development when machines become smart enough to make even smarter versions of themselves without human input.
He is talking rubbish, of course, but the story is the story here. Why would he say this? Why now?
Obviously concerned that this tidal wave of propaganda was being too subtle, in came NBC to really
hammer the point home:
The breaches signal that AI’s expanding capabilities are already fueling the security threat experts long feared.
Everybody got that?
It goes on and on.
Just today we’re
getting reports from the UK’s brand new
AI Security Institute that
rogue (there’s that word again) AI agents had tried to trick people by creating false identities.
And I think that’s beginning to circle around to at least part of the wider agenda here.
After all, if AI is so good at pretending to be human, and is out there creating fake identities to steal data, well…how can we tell who is
really human?
Perhaps everyone should be issued a digital ID to assist in
“proving humanity”.
That’s not me saying it, it’s think tanks like
OneID or “experts”
testifying in front of congress.
But the ever present push of Digital ID is just one of the more obvious of a raft of potential issues raised by the scary AI narrative.
For now we’re still in the generating fear phase. Getting pumped full of scare hormones by articles like this…
Misleading AI-generated doctors pose ‘huge danger to public safety’
- The Guardian
While they tease vague solutions in headlines
like this…
That’s a nice one, because it garnishes the AI main course with a soupçon of lab-leak flavour (something else that is doing the fear-porn rounds this week).
There are hints of legislation to come, the EU’s “AI transparency” rules just came into force this week, and the US demanding “kill switches” in AI models, but they are vague for now.
My soft prediction is that, before the end of the year, there will be a “rogue AI” false flag, which allegedly causes either loss of life or some massive financial losses.
I think financial would be the way to go, because they’re always looking for a reason to start a recession or banking crisis, and one they can blame on a third party is too good an opportunity to miss.
Whatever happens, there will be loud calls to “regulate AI”, but what that might mean in real terms I couldn’t say at this point.
To be clear though, if and/or when the official solutions to made-up problem #4987 are finally rolled out, they will have nothing at all do with controlling AI, and everything to do with controlling you.
