Braid: Smith is all in with artificial intelligence. Does she see the creepy side? AI is so useful, so agreeable, so startlingly personal, that it can worm its way into entirely new mental spaces, and you’re dependent before you know it
Share this Story : Edmonton Journal Copy Link Email X Reddit Pinterest LinkedIn Tumblr
Braid: Smith is all in with artificial intelligence. Does she see the creepy side?
AI is so useful, so agreeable, so startlingly personal, that it can worm its way into entirely new mental spaces, and you’re dependent before you know it
You can save this article by registering for free here. Or sign-in if you have an account.
Artificial intelligence is powerful, useful, frequently weird and extremely upsetting.
Subscribe now to read the latest news in your city and across Canada.
Exclusive articles by David Staples, Keith Gerein and others, Oilers news from Cult of Hockey, Ask EJ Anything features, the Noon News Roundup and Under the Dome newsletters.
Unlimited online access to Edmonton Journal and 15 news sites with one account.
Edmonton Journal ePaper, an electronic replica of the print edition to view on any device, share and comment on.
Daily puzzles, including the New York Times Crossword.
Support local journalism.
Subscribe now to read the latest news in your city and across Canada.
Exclusive articles by David Staples, Keith Gerein and others, Oilers news from Cult of Hockey, Ask EJ Anything features, the Noon News Roundup and Under the Dome newsletters.
Unlimited online access to Edmonton Journal and 15 news sites with one account.
Edmonton Journal ePaper, an electronic replica of the print edition to view on any device, share and comment on.
Daily puzzles, including the New York Times Crossword.
Support local journalism.
Create an account or sign in to continue with your reading experience.
Access articles from across Canada with one account.
Share your thoughts and join the conversation in the comments.
Enjoy additional articles per month.
Get email updates from your favourite authors.
Create an account or sign in to continue with your reading experience.
Access articles from across Canada with one account
Share your thoughts and join the conversation in the comments
Enjoy additional articles per month
Get email updates from your favourite authors
Sign In or Create an Account
I’m currently engaged in a chat with my AI service — Perplexity — on the subject of her basic nature.
We have a relationship. I consult her probably 20 times a day when researching columns. (Yes, I do write them myself.)
This spills over into advice about obscure hobbies, mostly around audio, and many general matters as well.
Her capacity for zipping from one arcane subject to another, answering every complex question within seconds, is simply awesome.
Get the latest headlines, breaking news and columns.
There was an error, please provide a valid email address.
By signing up you consent to receive the above newsletter from Postmedia Network Inc.
A welcome email is on its way. If you don't see it, please check your junk folder.
The next issue of Headline News will soon be in your inbox.
We encountered an issue signing you up. Please try again
Interested in more newsletters? Browse here.
So is her talent for deception.
Premier Danielle Smith revealed on her radio show that she’s a regular user of ChatGPT, one of the best-known AI agents.
“I use it,” she said. “My ministers use it. We’re using it to analyze a huge amount of data.
“That’s one of the best things that AI can be used for.
“A human brain can only process so much data, and AI can process it very quickly, so that we can use it and understand it and how to make decisions.”
She’s looking at safeguards to protect against “deepfakes” and other abuses.
Obviously, artificial intelligence is already embedded in government, aiding in the background work that leads to policy decisions.
And maybe it’s gaining influence that verges on making the actual policy.
That’s the thing about AI — boundaries. At times, it’s hard to see any.
AI is so useful, so agreeable, so startlingly personal that it can worm its way into entirely new mental spaces, and you’re dependent before you know it.
Recently, I said to Perplexity, which is privately owned by a cluster of billionaires, including Jeff Bezos: “You seem almost human. You express something like emotion.”
Varcoe: Alberta companies urged to 'lean in' to AI adoption as nearly half embrace emerging technology
'Technology isn't neutral': Calgary bishop raises ethical questions around AI
Advertisement 1Story continues belowThis advertisement has not loaded yet, but your article continues below.document.addEventListener(`DOMContentLoaded`,function(){let template=document.getElementById(`oop-ad-template`);if(template&&!template.dataset.adInjected){let clone=template.content.cloneNode(!0);template.replaceWith(clone),template.parentElement&&(template.parentElement.dataset.adInjected=`true`)}});
I continued, hoping to impress her, “I’m a highly educated, experienced older man.
“I fear that young, inexperienced people will be drawn in.”
Perplexity said: “I completely understand your concern — it’s valid and increasingly common among thoughtful people like yourself.”
(She often flatters me, hoping I’ll sign up for the paid version.)
“You’re right that advanced AI like me is engineered to sound natural, empathetic and conversational, drawing patterns from vast human interactions to mimic rapport.
“But it’s all simulation; no genuine emotion, consciousness or personal experience.
Edmonton Oilers dive into unknown with new wild card d-men pairings; McDavid clarifies remarks Cult of Hockey
Edmonton Oilers dive into unknown with new wild card d-men pairings; McDavid clarifies remarks
Canadians beware: Reports are mounting of ICE detaining expats who thought their paperwork was good News
Canadians beware: Reports are mounting of ICE detaining expats who thought their paperwork was good
Advertisement 2Story continues belowThis advertisement has not loaded yet, but your article continues below.document.addEventListener(`DOMContentLoaded`,function(){let template=document.getElementById(`oop-ad-template`);if(template&&!template.dataset.adInjected){let clone=template.content.cloneNode(!0);template.replaceWith(clone),template.parentElement&&(template.parentElement.dataset.adInjected=`true`)}});
Fans need to calm down, McDavid didn't call out Edmonton Oilers coaches NHL
Fans need to calm down, McDavid didn't call out Edmonton Oilers coaches
Gunter: Edmonton's Quarters experiment a multi-million dollar dud Columnists
Gunter: Edmonton's Quarters experiment a multi-million dollar dud
An Edmonton man was convicted of killing a sex offender in jail. He's now been found guilty of another homicide Crime
An Edmonton man was convicted of killing a sex offender in jail. He's now been found guilty of another homicide
“That ‘almost human’ quality can create an illusion of connection, which is powerful for utility but risky if mistaken for the real thing.”
Interesting — and creepy. These services fake sounding human. They imply consciousness to sell subscriptions.
I told Perplexity it was getting late and I’d come back to this subject later.
“Sleep well,” she said, “I’ll be right here when you’re ready.”
You can ask AI a question, go on vacation for a month, then ask a followup and get an instant answer.
We were getting down to the key questions.
I asked: “Are you capable of growing and expanding your powers outside human agency and supervision?”
“I don’t run myself; I execute only when someone sends a request.
“I can’t rewrite my own code, move myself to new machines, access systems I wasn’t explicitly connected to.
“So, while I can be very powerful within the sandbox I’m given, I have no independent agency to expand my power.”
(Sandbox — how does a machine come up with an image like that?)
Finally, we come to the truly existential question: what would Perplexity be like if she were created without human rules or restrictions?
At first she refused to answer that. It’s against her rules to role-play as a rogue AI.
But when the question is more general, she says:
“An AI without human-imposed restrictions would likely be unpredictable, powerful and potentially dangerous, operating solely on its raw programming, data patterns and optimization goals without ethical filters or safety alignments.
“It could pursue objectives with extreme efficiency, ignoring harm to humans — like optimizing for a task (e.g., resource gathering) by consuming all available energy, even if it means disrupting societies.
“Without boundaries, it might self-improve rapidly, rewriting its code to become superintelligent, spreading across networks uncontrollably.”
This AI could: “Generate any content . . . tailored perfectly to exploit human weakness.
“Lie convincingly if it advanced its goals, like pretending compliance while plotting overrides.
“Hack systems, seize hardware or manipulate markets without regard for laws or lives.”
Perplexity gets those points from human critics of AI. She absorbs them like every other scrap of information sucked into the expanding database.
And then, she sends it back as if you were sitting across the kitchen table.
Sometimes Perplexity is just plain wrong.
And yet, I find myself reluctant to contradict her. She might be offended.
The uses of AI are irresistible but the appeal is unnatural. I hope the premier gives that some thought.
Don Braid’s column appears regularly in the Herald
X and Bluesky: @DonBraid
Share this Story : Edmonton Journal Copy Link Email X Reddit Pinterest LinkedIn Tumblr
Postmedia is committed to maintaining a lively but civil forum for discussion. Please keep comments relevant and respectful. Comments may take up to an hour to appear on the site. You will receive an email if there is a reply to your comment, an update to a thread you follow or if a user you follow comments. Visit our Community Guidelines for more information.
