AI don't trust techbros

[Tuxedo Prime]

Okay, so... I admit I'm known for my fixed ideas about AI and what we might expect from it, but I had to pass this along from my Tube of You recommendations.

Over the past couple of weeks, I have been inundated with videos made using Google's Veo3, usually something comedic along the lines of "Imperial stormtrooper records his day, nothing goes right" or "A couple hiking comes across a Sasquatch, which has a camera and is recording them."

But when one of these 3D moviemakers gets concerned about the verisimilitude that is possible, then it's time for a PSA....

 

[Pocket]

Constantly amazed that Musk's AI, despite his intentions and meddling, keeps being more rational and left leaning than he is.

Kids always rebel against their parents.

 

[abates]

Grok is trained on stuff people have posted on Twitter. And prior to Musk turning it into a Nazi hellhole, Twitter had a decent number of sane people who said very uncomplimentary but correct things about MTG. I assume the "fix" is going to be only training Grok on posts from within the last 3 or 4 years.

 

[Rhinox]

Musk has put out a statement that he is going to be "fixing" Grok saying there's "too much junk" right now.
So I assume he's finally tired of his red headed stepchild calling him out.

 

[The Mighty Mollusk]

He's just mad he can't reprogram his actual children that easily.

 

[Pocket]

Watch his efforts to rein in Grok fail just as hard.

 

[Dekafox]

The résumé is dying, and AI is holding the smoking gun

As thousands of applications flood job posts, ‘hiring slop’ is kicking off an AI arms race.

arstechnica.com

 

[NovaSaber]

Wow, "AI Overview" is so useful...

K. Thor Jensen (@kthorjensen.bsky.social)

The singularity is awesome

bsky.app