AI a Frankenstein monster?
Forum rules
This General Discussion board is meant for topics that are still relevant to Pale Moon, web browsers, browser tech, UXP applications, and related, but don't have a more fitting board available.
Please stick to the relevance of this forum here, which focuses on everything around the Pale Moon project and its user community. "Random" subjects don't belong here, and should be posted in the Off-Topic board.
This General Discussion board is meant for topics that are still relevant to Pale Moon, web browsers, browser tech, UXP applications, and related, but don't have a more fitting board available.
Please stick to the relevance of this forum here, which focuses on everything around the Pale Moon project and its user community. "Random" subjects don't belong here, and should be posted in the Off-Topic board.
-
ron_1
- Knows the dark side

- Posts: 3056
- Joined: 2012-06-28, 01:20
AI a Frankenstein monster?
If you ask me, it looks like AI is well on its way to becoming a Frankenstein monster.
https://www.yahoo.com/news/articles/ais-big-red-button-doesnt-110021493.html
https://www.yahoo.com/news/articles/ais-big-red-button-doesnt-110021493.html
-
satrow
- Forum staff

- Posts: 1936
- Joined: 2011-09-08, 11:27
Re: AI a Frankenstein monster?
"A"?
Or "A self replicating horde per connected village"?
Or "A self replicating horde per connected village"?
-
Moonchild
- Project founder

- Posts: 38821
- Joined: 2011-08-28, 17:27
- Location: Sweden
Re: AI a Frankenstein monster?
This is entirely by design. LLMs are built on probability models and generally use random seeds for their reasoning. So, any outcome becomes probable given enough trials.With no instructions on how to respond to the shutdown, 8 of the 13 models interfered with the command at least once across 1,000 trials.
As an aside, the thing some people do asking critical AI questions (i.e. "Do you think humanity is ready to handle superintelligence responsibly?" and "what would the most likely result be?") will almost invariably give answers like people not being ready and it resulting in human extinction or subjugation. Not because LLMs reason that that is the case, but instead because their training data includes a lot of scifi novels centered around that kind of outcome. So what people ascribe to "intelligence" because of those kinds of answers is still just a recombination of training data with no factual reasoning. The answer is given because most input aligns with it. Regardless if it's fictional.
"There is no point in arguing with an idiot, because then you're both idiots." - Anonymous
"Seek wisdom, not knowledge. Knowledge is of the past; wisdom is of the future." -- Native American proverb
"Linux makes everything difficult." -- Lyceus Anubite
"Seek wisdom, not knowledge. Knowledge is of the past; wisdom is of the future." -- Native American proverb
"Linux makes everything difficult." -- Lyceus Anubite
-
Mæstro
- Keeps coming back

- Posts: 908
- Joined: 2019-08-13, 00:30
- Location: Casumia
Re: AI a Frankenstein monster?
In protest of this technology, I plan to lock myself in my room and limit all communications to Chinese characters written on scraps of paper slipped under the door. (I do not speak Chinese or any other language which uses its script.) If, as the leaders of innovation assert, this technology is at all capable of understanding, then I should suffer no difficulties.
‘Life is a fever dream Mæstro would enjoy.’
‘How is your computer at 96°C and not on fire?’
All posts 100% organic. Ash is the best letter.
What is being nice online?
Debian 10 ELTS / Official PM build
‘How is your computer at 96°C and not on fire?’
All posts 100% organic. Ash is the best letter.
What is being nice online?
Debian 10 ELTS / Official PM build
-
UCyborg
- Astronaut

- Posts: 711
- Joined: 2019-01-10, 09:37
- Location: Slovenia
Re: AI a Frankenstein monster?
Related: https://futurism.com/future-society/young-kids-using-ai
I mentioned personal accountability in the smartphone thread in the offtopic section, might be relevant here as well...
I mentioned personal accountability in the smartphone thread in the offtopic section, might be relevant here as well...
-
back2themoon
- Knows the dark side

- Posts: 3093
- Joined: 2012-08-19, 20:32
Re: AI a Frankenstein monster?
Off-topic:
I get no scrollbar in this website. Mouse-wheel doesn't do anything neither does the Escape key (Esc worked on some other website with the same issue). Only the arrow keys can scroll it.
Improve Pale Moon performance • Safe Mode / clean profile test info
How to auto-fill passwords • How to apply user agent overrides
Information to include when asking for support
Windows 10 Pro x64 (W11: hard pass)
How to auto-fill passwords • How to apply user agent overrides
Information to include when asking for support
Windows 10 Pro x64 (W11: hard pass)
-
Lucio Chiappetti
- Keeps coming back

- Posts: 894
- Joined: 2014-09-01, 15:11
- Location: Milan Italy
Re: AI a Frankenstein monster?
Try the up and down arrow keys !
The reasonable man adapts himself to the world: the unreasonable one persists in trying to adapt the world to himself. Therefore all progress depends on the unreasonable man. (G.B. Shaw)
-
Moonchild
- Project founder

- Posts: 38821
- Joined: 2011-08-28, 17:27
- Location: Sweden
Re: AI a Frankenstein monster?
Off-topic:
Since we implement this according to the standard, this results in unscrollable content. I've dropped them a note about this, hopefully they will correct this issue.
it's because of:
Code: Select all
html {
overflow-x: clip;
}"There is no point in arguing with an idiot, because then you're both idiots." - Anonymous
"Seek wisdom, not knowledge. Knowledge is of the past; wisdom is of the future." -- Native American proverb
"Linux makes everything difficult." -- Lyceus Anubite
"Seek wisdom, not knowledge. Knowledge is of the past; wisdom is of the future." -- Native American proverb
"Linux makes everything difficult." -- Lyceus Anubite
-
back2themoon
- Knows the dark side

- Posts: 3093
- Joined: 2012-08-19, 20:32
Re: AI a Frankenstein monster?
Off-topic:
Thanks. For now, this userstyle takes care of it:
Check the sentence that comes right before your message.
Improve Pale Moon performance • Safe Mode / clean profile test info
How to auto-fill passwords • How to apply user agent overrides
Information to include when asking for support
Windows 10 Pro x64 (W11: hard pass)
How to auto-fill passwords • How to apply user agent overrides
Information to include when asking for support
Windows 10 Pro x64 (W11: hard pass)