While putting an AI computer in direct charge of life-extinction lethal
force is an obvious no-no to anyone familiar with sci-fi doomsday
scenarios, it doesn't take extremes like Colossus to cause potential
disaster.
If humans ever get so dependent on AI that they blindly follow the
"advice" of an AI assistant, all it takes is for AI to feed humans a
mis-analysis of a life-threatening situation or potential extinction
event and let the humans do the damage. This doesn't require that the
AI be sentient or self-preserving -- only that its advice be assumed
valid and followed in situations where the AI is not capable of making
good judgements or correctly assessing the consequences for humanity. I
see this as a more imminent danger than the future possibility of AI
becoming sentient and deciding humans are a problem best eliminated.
As progress continues to be made on making AI increasingly complex and
getting AI to emulate more and more of human behavior, it would also be
presumptuous to assume that that we might not at some point accidentally
construct an AI device capable of emulating bad human traits, like
sociopathic or psychopathic behavior -- and it might not be obvious as
humans with these issues sometimes hide it well. This doesn't require
the AI to be sentient to do damage, only that it give advice that is
subtly not in the best interest of humanity.
I see no obvious way to stop AI "progress", but constant vigilance
against abuse is obviously warranted.
Joel C. Ewing
On 4/10/23 08:50, Bob Bridges wrote:
Or, of course, "Colossus: The Forbin Project", in which the computer
concludes that it must become the world's dictator in order to preserve
mankind from its own idiocies.
---
Bob Bridges, [email protected], cell 336 382-7313
/* Failure is the opportunity to begin again more intelligently. -ok, I
found it in a fortune cookie, but it's not bad */
-----Original Message-----
From: IBM Mainframe Discussion List <[email protected]> On Behalf Of
Seymour J Metz
Sent: Monday, April 10, 2023 08:52
Or the chilling "The Humanoids" by Jack Williamson "To serve Man, and keep
him from harm".
________________________________________
From: Allan Staller <[email protected]>
Sent: Monday, April 10, 2023 8:23 AM
Why does the author omit the classic "I, Robot" by Issac Asimov.
Where the "AI" decides to protect humanity from itself.
That being said, I believe we are far from sentient AI at this point in
time.
-----Original Message-----
From: IBM Mainframe Discussion List <[email protected]> On Behalf Of
Bill Johnson
Sent: Sunday, April 9, 2023 2:01 PM
The Aliens Have Landed, and We Created Them
https://www.bloomberg.com/opinion/articles/2023-04-09/artificial-intelligenc
e-the-aliens-have-landed-and-we-created-them
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN
--
Joel C. Ewing
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN
