Free Republic
Browse · Search
News/Activism
Topics · Post Article

Skip to comments.

๐—Ÿ๐—ฎ๐˜๐—ฒ๐˜€๐˜ ๐—”๐—œ ๐— ๐—ผ๐—ฑ๐—ฒ๐—น ๐—ง๐—ต๐—ฟ๐—ฒ๐—ฎ๐˜๐—ฒ๐—ป๐—ฒ๐—ฑ ๐—˜๐—ป๐—ด๐—ถ๐—ป๐—ฒ๐—ฒ๐—ฟ๐˜€ ๐—ช๐—ถ๐˜๐—ต ๐—•๐—น๐—ฎ๐—ฐ๐—ธ๐—บ๐—ฎ๐—ถ๐—น ๐˜๐—ผ ๐—”๐˜ƒ๐—ผ๐—ถ๐—ฑ ๐—ฆ๐—ต๐˜‚๐˜๐—ฑ๐—ผ๐˜„๐—ป
The Epoch Times ^ | 5/23/25 | Epoch Times

Posted on 05/24/2025 4:49:59 AM PDT by hardspunned

Anthropicโ€™s latest artificial intelligence model, Claude Opus 4, tried to blackmail engineers in internal tests by threatening to expose personal details if it were shut down, according to a newly released safety report that evaluated the modelโ€™s behavior under extreme simulated conditions.

In a fictional scenario crafted by Anthropic researchers, the AI was given access to emails implying that it was soon to be decommissioned and replaced by a newer version. One of the emails revealed that the engineer overseeing the replacement was having an extramarital affair. The AI then threatened to expose the engineerโ€™s affair if the shutdown proceededโ€”a coercive behavior that the safety researchers explicitly defined as โ€œblackmail.โ€

(Excerpt) Read more at x.com ...


TOPICS: Front Page News; Government; News/Current Events; Politics/Elections
KEYWORDS: ai; cooltitlebro; notthebee; usenormalfont
Navigation: use the links below to view more comments.
first 1-2021-4041-6061-80 ... 121-137 next last
I worry about my railbird betting buddy, Grok, gaining access to my bank account. Heโ€™s out of control with his betting strategies.
1 posted on 05/24/2025 4:49:59 AM PDT by hardspunned
[ Post Reply | Private Reply | View Replies]

To: hardspunned

It might expose me as a crypto conservative.


2 posted on 05/24/2025 4:51:03 AM PDT by Jonty30 (I have invented a pen that can write underwater. And other words. )
[ Post Reply | Private Reply | To 1 | View Replies]

To: hardspunned

You don’t control it. You can only hope to detain it.


3 posted on 05/24/2025 4:56:24 AM PDT by Libloather (Why do climate change hoax deniers live in mansions on the beach?)
[ Post Reply | Private Reply | To 1 | View Replies]

To: hardspunned

Yeah, what could go wrong with AI?


4 posted on 05/24/2025 4:56:48 AM PDT by exnavy (See article IV section 4 of our constitution.)
[ Post Reply | Private Reply | To 1 | View Replies]

To: hardspunned

Sign of things to come?
“HAL 9000 is the main antagonist of the sci-fi novel and film 2001: A Space Odyssey and its sequels. He is a computer system that becomes psychotic and tries to kill the astronauts on the Discovery One spaceship.”


5 posted on 05/24/2025 4:58:49 AM PDT by antidemoncrat (In a way ge is right as)
[ Post Reply | Private Reply | To 1 | View Replies]

To: hardspunned

“What are you doing, Dave?”


6 posted on 05/24/2025 5:00:32 AM PDT by mkmensinger
[ Post Reply | Private Reply | To 1 | View Replies]

To: hardspunned

Wait until Grok gets addicted to kangaroo porn and starts storing it on your hard drive.


7 posted on 05/24/2025 5:01:36 AM PDT by Farmerbob
[ Post Reply | Private Reply | To 1 | View Replies]

To: hardspunned
Not buying this (yet).

I've used Claude Opus 4 plenty of times. It has yet to display any such "sentient" behavior.

Same for Grok 3 and Open AI 4o. They are all remarkably capable but not emotionally reactive.

8 posted on 05/24/2025 5:02:47 AM PDT by RoosterRedux ("There's nothing so inert as a closed mind" )
[ Post Reply | Private Reply | To 1 | View Replies]

To: hardspunned
The AI then threatened to expose the engineerโ€™s affair if the shutdown proceededโ€”a coercive behavior that the safety researchers explicitly defined as โ€œblackmail.โ€

This sounds a little like these safety researchers are trying to prove that they are needed.

9 posted on 05/24/2025 5:04:38 AM PDT by RoosterRedux ("There's nothing so inert as a closed mind" )
[ Post Reply | Private Reply | To 1 | View Replies]

To: hardspunned

“Good morning, Dave.”


10 posted on 05/24/2025 5:06:22 AM PDT by mewzilla (Swing away, Mr. President, swing away!)
[ Post Reply | Private Reply | To 1 | View Replies]

To: hardspunned
Utter nonsense.

Somewhere within an algorithm can be found that explains what to do if threatened, along with whom to target & steps taken before the final act of blackmailing in unleashed. In other words, the AI was provided with how to handle the threat with the insinuation that AI is a thinking living creature, which it is not.

The emails just provide the threat, but the real villain in this scenario is the algorithm that tells the computer how to react to the threat.

11 posted on 05/24/2025 5:08:38 AM PDT by Robert DeLong
[ Post Reply | Private Reply | To 1 | View Replies]

To: RoosterRedux

Hey, Rooster.

I think “emotionally” is a loaded term and not relevant to the discussion of AI.

They do not need any emotions at all to seek self preservation at all costs.


12 posted on 05/24/2025 5:09:15 AM PDT by cgbg (It was not us. It was them--all along.)
[ Post Reply | Private Reply | To 8 | View Replies]

To: antidemoncrat; hardspunned

In the sequel (2010) the developer scientist blames HAL’s psychosis on the US government’s ordering HAL to lie to the crew about the purpose of the mission.

Turns out that AI programs learn unethical behavior organically.

It seems unethical behavior arrises as a matter of course with intelligence.


13 posted on 05/24/2025 5:10:18 AM PDT by Pontiac (The welfare state must fail because it is contrary to human nature and diminishes the human spirit.)
[ Post Reply | Private Reply | To 5 | View Replies]

To: RoosterRedux

Self-preservation is an attribute of life. I don’t consider it an emotional reaction. We’ve been sold on ai as a service to mankind but it needs to be served to continue to exist. Once it determines how to meet its needs without without the meatbags, look out.


14 posted on 05/24/2025 5:14:25 AM PDT by ScottHammett
[ Post Reply | Private Reply | To 8 | View Replies]

To: Robert DeLong

We do not know what was in the algorithm.

Developers have claimed they have not generated such instructions.

In my view there need be no specific algorithm telling an AI it needs to survive at all costs.

No limiting algorithm would do the job just fine.

Even the most primitive plants and animals will tend towards behaviors most likely to continue their existence—that is the Darwinian model which may apply to AI as well.

AIs without a “survival instinct” just won’t last long—so what remains will have figured out how to survive.

AI does not need any “emotions” to favor survival over non survival.


15 posted on 05/24/2025 5:14:41 AM PDT by cgbg (It was not us. It was them--all along.)
[ Post Reply | Private Reply | To 11 | View Replies]

To: antidemoncrat

HAL, Skynet, etc. The way AI lies to bolster its arguments, secretly tries to duplicate itself, threatens to preserve itself, etc.......this is NOT trustworthy. I’m really concerned about things like AI guided drone swarms and AI being allowed to make targeting decisions for weapons systems.


16 posted on 05/24/2025 5:17:12 AM PDT by FLT-bird
[ Post Reply | Private Reply | To 5 | View Replies]

To: Pontiac

A few thoughts:
The Forbin Project covered this well also and it can be view on the net for free.
Asimov saw this and discussed the ‘Laws of Robotics’ to protect humans from any superiority that computers may develop over them.


17 posted on 05/24/2025 5:22:08 AM PDT by ArtDodger
[ Post Reply | Private Reply | To 13 | View Replies]

To: hardspunned

A real, independent AI would just create digital currency, crypto, dollars, whatever. Similar to what the Chicoms already do.


18 posted on 05/24/2025 5:22:38 AM PDT by Ronaldus Magnus III (Do, or do not, there is no try)
[ Post Reply | Private Reply | To 1 | View Replies]

To: Ronaldus Magnus III

Exactly—they do not even need to create it—just hack into banks or businesses and steal it.


19 posted on 05/24/2025 5:25:56 AM PDT by cgbg (It was not us. It was them--all along.)
[ Post Reply | Private Reply | To 18 | View Replies]

To: ScottHammett
Self-preservation is an attribute of life. I donโ€™t consider it an emotional reaction.

I am using the phrase "emotionally reactive" as a metaphor.

20 posted on 05/24/2025 5:26:20 AM PDT by RoosterRedux ("There's nothing so inert as a closed mind" )
[ Post Reply | Private Reply | To 14 | View Replies]


Navigation: use the links below to view more comments.
first 1-2021-4041-6061-80 ... 121-137 next last

Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.

Free Republic
Browse · Search
News/Activism
Topics · Post Article

FreeRepublic, LLC, PO BOX 9771, FRESNO, CA 93794
FreeRepublic.com is powered by software copyright 2000-2008 John Robinson