ChatGPT

Botch

For Sale: Baby shoes.  Never worn.
 
Dayam.  (60 Minutes, CBS)  

nolaegghead

Saw that, interesting 

Botch

I'm not a big movie fan, but the films of Stanley Kubrick have impacted me, at least those that I've seen.  Got a reminder of this incredible scene from one of them, tonight.
 
https://youtu.be/dSIKBliboIo
 
Interesting that, in light of what ChatGPT can do now, HAL let David Bowman ask essentially the same question, many times, before finally responding, giving him a "Haha, caught you!" sort of statement, before again cutting him off (in this scene).
 
Bowman managed to get back into the ship, and then slowly dismembered HAL's "cognitive" brain, one Allen* screw at a time (HAL decided to restart the conversation and (iirc) David then wouldn't respond to HAL; fascinating to me at this time!)  I need to dig this disk out and watch it again tonight; one of my all-time favorite movies.  
 
Daisy, Daisy, ....  
 
 
*it may have been a Phillips  

lousubcap

A dirrerent perspective regarding ChatGPT and the classroom:

https://www.technologyreview.com/2023/04/06/1071059/chatgpt-change-not-destroy-education-openai/?utm_

lousubcap

Some more regarding AI:

"When you see something that is technically sweet, you go ahead and do it and you argue about what to do about it only after you have had your technical success." So said Robert Oppenheimer about the development of an atomic weapon. That was a quote often paraphrased by Geoffrey Hinton when discussing his role in another arms race. But the man known as "The Godfather of AI" has soured on the sweet tech he helped create. Hinton is worried about the risks his project now poses to humanity, in part because, as he explains, "It is hard to see how you can prevent the bad actors from using it for bad things." That concern wasn't raised by insiders thirty years ago, when the world wide web launched into in the public domain, and we're the worse for our over-exuberance at the expense of measured development. But the web's development was a slow moving locomotive compared to AI's self-acceleration that's already faster than a speeding bullet. NYT (Gift Article): ‘The Godfather of A.I.' Leaves Google and Warns of Danger Ahead. Hinton now looks back and says, "I console myself with the normal excuse: If I hadn't done it, somebody else would have."

Edit in an attempt to clean up spacing.

Botch

https://youtu.be/-eAQOhDNLt4 

Rick Beato has some interesting thoughts on what AI-generated "songs" can/will do to the Music Industry here.  He concludes that the Industry will eventually cut out the Artist altogether, but imnsho he doesn't take it far enough: These deepfakes weren't even generated by the "Music Industry", they were generated by some kid in their Mommy's basement!  

This tech is already disrupting the music, photography, and politics industries, and once the video portions get better (past the beer and pizza commercials) it will be disrupting TV (both nat'l and local), Hollywood, and even professional sports.  We live in Interesting Times.  

nolaegghead

lousubcap

From Fareed Zakaria:

How Authoritarians Could Use AI

"AI is moving fast. Just a few months after prompting much hype, artificial intelligence (AI) chatbot ChatGPT is “already starting to look outdated,” Matteo Wong writes for The Atlantic; “multimodal” AI models (which train on more than just text, including images and audio) are doing more.     In recent newsletters, we’ve explored the debate over how to regulate AI and whether China can catch up, after US-developed ChatGPT took an early lead. Relatedly, observers have warned for years that emerging technologies can be used by authoritarian regimes (like China’s) to maintain control.     How might those regimes use AI? Author of “The Rise of Digital Repression: How Technology Is Reshaping Power, Politics, and Resistance,” Steven Feldstein tells J.J. Gould in a recent interview for The Signal newsletter that “a few use cases stand out …  One is tracking popular discontent and, when it comes to it, controlling mass protest. That can work in a number of different ways. It can work on a mass scale through automated social-media monitoring, interpreting what people are thinking from what they’re saying online. It can work through public-surveillance cameras and other ways of seeing when and where people are gathering—and then preempting political demonstrations or arresting people who participate in them. … A second use case is maintaining control in an area of a country where the state is experiencing unrest. … A third … is super-enhancing propaganda and disinformation.”    As China and other countries innovate, Feldstein warns that repressive, AI-enhanced technologies can proliferate quickly—“much more like conventional arms” than nuclear weapons—and that democracies must regulate AI and solidify norms surrounding it sooner than later." Edit in an attempt to clean up spacing.

Legume

Has NRA issued guidance yet on how many guns are needed to protect against AI taking my freedom?

JohnInCarolina

lousubcap

Skilled trades will always be the backbone of infrastructure.  In my limited sphere of reading, the younger generation gets this as the trades. coupled with the exorbitant cost of college are drivers. 

Botch

JohnInCarolina said:

I left Thiokol in 1999, and the next spring I had to re-do my entire sprinkling system.  Hired a guy with good ratings, recognized him as a tech up at Thiokol.  He worked his butt off all day Saturday and Sunday, system worked beautifully, and I got charged about as much as I made at Thiokol in two weeks, as a Senior Engineer.  (I still hire him for minor repairs, if needed, he's good)
Any youngster that asks me for career advice, I tell them "Do something with your hands, that can't be sourced overseas (and now include AI outsourcing, although that's hard to determine at this point)".  Still curious as to how this will be affecting computer programmers, over the next few years.  
 

dbCooper

"Have 10 hours? IBM will train you in AI fundamentals - for free"

In a different life I was sold off to IBM in a outsourcing deal, worked for them about five years.  There were many negatives associated with the bureaucracy of the company, which was my main driver in leaving them.  The positives included the vast knowledge gained from the people I worked with and the excellent in-house educational resources available. 

If you or someone you know have an interest in AI fundamentals, I'd give this a look... https://www.zdnet.com/article/have-10-hours-ibm-will-train-you-in-ai-fundamentals-for-free/

I plan to take the course and fully expect it to be time well spent.

xavier233

ChatGPT is not the only AI tool out there. So many of them are available now that it's almost scary.
Like, here's a list of AI writing tools: https://writingtools.co.uk/pricing.html. I mean, wow, just look how many. Afraid to think what the future holds for us.

Canugghead

Speaking of skilled trades, just heard this story yesterday...at one of our kids' office there was a funny noise coming from the false ceiling, they discovered a small fan that had been left running for over a decade by a contractor.