Site Updates AI Question?

 

[Lank72]

My philisophy on life in numbers:

1. Thinking of a loan? Don't get a loan
2. Don't get a loan
3. STEP AWAY FROM THE LOAN

Save own money first, get the shiny thing later

 

[node]

I know an architect, prob like you (@Muiredach ) but much older and you can see plenty of comments over at theregister.com - he said its quite a disaster really, hallucinating and all other stupid stuff, which could be dangerous and/or have liability implications in a clinical environment. Honestly I'd avoid it like the plague.
To add, it is, as usual, the big money people who are mostly related to each other behind this, so they win multiple times stealing more potentially PII and selling the vastly overpriced hardware,on and on.. Guess what? Same for the energy companies... In my view its yet another racket (or better word grift) from our well connected terribles playing all sides (in this case the markets). Business as usual.
Tech explanation for our users : https://en.wikipedia.org/wiki/Hallucination_(artificial_intelligence) PII: https://en.wikipedia.org/wiki/Personal_data also don't be blinded by the term AI, its an https://en.wikipedia.org/wiki/Large_language_model, thats it. The only "intelligence" in it is exactly the kind you absolutely don't want - Trouble in other words - and some are saying this will offer, in time, our fat controllers Total control. so better to be your quietly brilliant, outlier self... However, if you feel you must adopt it then at least please give us users a big OPT OUT button.

TLDR : I feel both morally and ethically, the only winning move is: not to play.

 

[utopiate]

it would be fucking cool though lol so I get it. However, I'm not sure it would bring that level of value to the site just yet. The Ad revenue may bring in the necessary before you think so maybe have it as a median goal? Given the data set it would be exposed to I can understand it's potential and accompanying increase in revenue. I guess the time it would take would suggest you start now though so hmm...it could be a colossal earner. I appreciate being consulted but think it likely you need some professional input. Happy to support whatever you decide.

 

[node]

Its a security shit show according to those in attendence at CYBERUK, this week:

"Peter Garraghan – CEO of Mindgard and professor of distributed systems at Lancaster University – asked the CYBERUK audience for a show of hands: how many had banned generative AI in their organizations? Three hands went up.
"And how many, in your deepest of hearts, actually have a good grasp of the security risks involved in AI system controls, by a show of hands?"
Not a single hand was raised among the 200-strong, security-savvy crowd."
https://www.theregister.com/2025/05/14/cyberuk_ai_deployment_risks/

 

[BudGuy]

My philisophy on life in numbers:

1. Thinking of a loan? Don't get a loan
2. Don't get a loan
3. STEP AWAY FROM THE LOAN

Save own money first, get the shiny thing later

Depends on the interest rate on the loan and the appreciative/depreciative nature of what you buy with it.

Don't get a loan at 30% APR for a ferrari.

But absolutely get a loan whose interest rate is below the rate of inflation (they did exist at one point lol) to fund a home improvement that adds value or saves money in the long run, go for it.

Everything else exists on a spectrum between those two lol. So the real question is ... is an AI chatbot a ferrari, or a set of solar panels?

I think it might be a ferrari. Not all the information in the database is complete or up to date, and the review system isn't up and running yet, and without that the AI chat bot won't be much more useful than what we already have, and ChatGPT can already reference what's public facing from MedBud as well as leafly and everywhere else all at once.

I say spend the money improving what MedBud already is, then things like ChatGPT will reference it more often, and that'll drive the traffic here. Then you're sort of harnessing the power of the AI for free without shouldering any of the risk for it.

 

[node]

I know an architect, prob like you (@Muiredach ) but much older and you can see plenty of comments over at theregister.com - he said its quite a disaster really, hallucinating and all other stupid stuff, which could be dangerous and/or have liability implications in a clinical environment. Honestly I'd avoid it like the plague.
To add, it is, as usual, the big money people who are mostly related to each other behind this, so they win multiple times stealing more potentially PII and selling the vastly overpriced hardware,on and on.. Guess what? Same for the energy companies... In my view its yet another racket (or better word grift) from our well connected terribles playing all sides (in this case the markets). Business as usual.
Tech explanation for our users : https://en.wikipedia.org/wiki/Hallucination_(artificial_intelligence) PII: https://en.wikipedia.org/wiki/Personal_data also don't be blinded by the term AI, its an https://en.wikipedia.org/wiki/Large_language_model, thats it. The only "intelligence" in it is exactly the kind you absolutely don't want - Trouble in other words - and some are saying this will offer, in time, our fat controllers Total control. so better to be your quietly brilliant, outlier self... However, if you feel you must adopt it then at least please give us users a big OPT OUT button.

TLDR : I feel both morally and ethically, the only winning move is: not to play.

Its more important to me at least, and for sure the advertisers - for us to get the wiki & the forum fully tested (see above dark mode bug) fixed,retested across multiple mobile devices / browsers and desktop browsers secured as best as possible.

 

[node]

"The tech industry is generating a tsunami of AI slop, along with a few instances of true value. In the coding world, generative AI can make for useful coding assistants but are often less than helpful when expected to behave like an engineer."

Builder.ai coded itself into a corner – now it's bankrupt

Comment: When 'AI-powered' means 'mostly humans and bad decisions'

www.theregister.com

Also https://www.theregister.com/2025/05/21/llm_torture_tools/

 

[Moon]

This old saying has served my life well: "Never a lender or borrower be".