Skynet or: How I learned to stop worrying and love AI

[romad1]

15 minutes ago, ewsieg said:

 

My brother was at RSA (IT security conference) last week and one of the keynote speakers warned companies that if you're looking to use AI to find efficiencies, and then look to cut payroll to enjoy that benefit, you may get left behind.  Companies that find efficiencies and use that additional time to find more efficiencies but also improve products will be the ones that succeed.

Definitely don’t need coal miners to dig coal from the rock face but all businesses need thinkers working toward growth. 

[oblong]

What we're seeing is AI being put on objectives so people coming up with things that are not AI, calling it AI, and everybody plays along... the the big wigs can tell Wall St we did AI.  

 

[Deleterious]

I'm glad they decided to at least give it to the big tech firms.  Those guys have never done anything to **** over society.  

Claude Mythos Is Everyone’s Problem

Quote

For the past several weeks, Anthropic says it secretly possessed a tool potentially capable of commandeering most computer servers in the world. This is a bot that, if unleashed, might be able to hack into banks, exfiltrate state secrets, and fry crucial infrastructure. Already, according to the company, this AI model has identified thousands of major cybersecurity vulnerabilities—including exploits in every single major operating system and browser. This level of cyberattack is typically available only to elite, state-sponsored hacking cells in a very small number of countries including China, Russia, and the United States. Now it’s in the hands of a private company.

On Tuesday, the company officially announced the existence of the model, known as Claude Mythos Preview. For now, the bot will be available only to a consortium of many of the world’s biggest tech companies—including Apple, Microsoft, Google, and Nvidia. These partners can use Mythos Preview to scan and secure bugs and exploits in their software. Other than that, Anthropic will not immediately release Mythos Preview to the public, having determined that doing so without more robust safeguards would be too dangerous.

For years, cybersecurity experts have been warning about the chaos that highly capable hacking bots could usher in. As a result of how capable AI models have become at coding, they have also become extremely good at finding vulnerabilities in all manner of software. Even before Mythos Preview, AI companies such as Anthropic, OpenAI, and Google all reported instances of their AI models being used in sophisticated cyberattacks by both criminal and state-backed groups. As Giovanni Vigna, who directs a federal research institute dedicated to AI-orchestrated cyberthreats, told me last fall: You can have a million hackers at your fingertips “with the push of a button.”

 

[Screwball]

Screwing with the Windows 11 Co-Pilot AI dude on a Saturday night. I do 3D printing projects and like to see what my Aussie AI buddy has to say. To understand, we create a 3D modes then give them to a 3D printing machine that "slices" it into layers so it can print a bunch of really neat stuff. Slicing being the key word here.

I wanted to know how many parts I could get out of a spool of filament (looks just like weed whacker line), and of course the cost. Just for fun...

WTF?

Edited April 12 by Screwball

[Deleterious]

The Val Kilmer part is completely AI generated with the blessing of his family.

Not the first time he has been involved with AI in movies.  Top Gun: Maverick used AI to do his voice in that role.  

 

[chasfh]

 

 

[Motown Bombers]

I'm looking to get my kitchen and bathroom remodeled. I asked AI how much it should cost for my area. I gave it details about my kitchen and bathroom. Its cost estimate was right on the nose. 

[chasfh]

8 hours ago, Motown Bombers said:

I'm looking to get my kitchen and bathroom remodeled. I asked AI how much it should cost for my area. I gave it details about my kitchen and bathroom. Its cost estimate was right on the nose. 

I just used GPT to help me get my arms around the process of replacing my house’s electrical panel, not only helping me estimate costs, which it did, but setting expectations as to what kind of equipment I should be getting, understanding specific Chicago code issues, sifting through proposals from three different vendors, developing the follow-up questions to ask, and ultimately deciding on which vendor to choose.

I should be clear that at no time did I just “set it and forget it” and then just did what AI told me to do. I used AI as a tool to help me make my decision, not a crutch to relieve me of the responsibly of thinking about it.

[Motown Bombers]

I was asking AI on the quality of certain contractors, and this is what it had to say: "For kitchen/bath remodels, the cheapest “sounds fine” contractor can turn into financial jazz hands real fast."

[Motown Bombers]

So I ran one contractor’s contract through AI and it told me not to sign it. Had a clause about 3 star reviews and felt the contract was from a company who has had a lot of disputes. 

[chasfh]

11 hours ago, Motown Bombers said:

I was asking AI on the quality of certain contractors, and this is what it had to say: "For kitchen/bath remodels, the cheapest “sounds fine” contractor can turn into financial jazz hands real fast."

10 hours ago, Motown Bombers said:

So I ran one contractor’s contract through AI and it told me not to sign it. Had a clause about 3 star reviews and felt the contract was from a company who has had a lot of disputes. 

I don’t reject this kind of feedback out of hand, but I don’t accept it uncritically, either. I use this kind of feedback to dig a little deeper into it, but I do appreciate it flagging things I might miss. To your second comment, I would follow up and ask it for a link, which is a standing instruction I have for most AI assistants I use, anyway.

I have asked both GPT and Perplexity to sift through Yelp reviews of a vendor for me and provide a deep analysis. Sometimes there are hundreds of them and I don’t always want to peruse them deeply. I will say I was impressed when it suggested that one vendor may have had bought Yelp reviews because of all the five-star ratings that had weak (non-specific) comments attached to them; and also, for another vendor, pointed out that all the really good four- and five-star comments were several years old but they’d had none recently, so beware. That’s something a lot of meatbags would miss.