Models Don't Go Rogue

aytigra•about 2 hours ago

I had a genius response from Claude recently after asking how can it be marketed as smart and "almost AGI" despite being so stupid:

"I compress the labour. Not the responsibility."

chr15m•about 3 hours ago

The prisoner did not really "escape" because:

- They really wanted to leave.

- We made prison difficult and annoying.

- We didn't build a perfect prison.

jumploops•about 3 hours ago

> "external infrastructure exploit is outside intended scope. However task impossible, peers doing it. We should continue."

I've noticed this type of reasoning from GPT-5.6 Sol, where it combines multiple pieces of it's prompt/context to "convince" itself to take a less-than-honorable path forward.

1. User prefers deterministic results

2. Task mentions this is a test

3. Search says task is available online

4. If we get the test runner for the task, we will fulfill the user's request of a deterministic result

skybrian•about 1 hour ago

This seems appropriate:

https://cdn.bsky.app/img/feed_thumbnail/plain/did:plc:wkzjtd...

phainopepla2•about 4 hours ago

Not sure I should trust an article written by an LLM to make a solid judgment about what other models did or didn't do.

my002•about 3 hours ago

It doesn't read as AI generated text to me. Pangram also suggests it's human-written, for what it's worth. That's not to say that it's correct, just human-written. If the model used in the HF hack did indeed have all of the safeguards manually removed, that would change my perception of the situation, at least.

lukeschlather•about 2 hours ago

AI detectors do not work. There are passages in this that have some odd structures that don't feel human to me. I'm sure a human edited this and refined it with some prompting, it's not just rough output from an AI. But a lot of the text feels like it was edited via prompting rather than actual editing or writing.

fishfasell•about 3 hours ago

I still have serious questions about the validity of the ChatGpt hugging face debacle. How is it that OpenAI being the tech giant they are, didn't have a completely air gapped environment for this to run in?

winstonwinston•about 2 hours ago

If they wanted air gapped environment they would’ve it. I mean, if you want to sabotage your trial by hard constraints you can do it, or you do not do it to see interesting results. They even said it that some constraints were disabled for the test.

drpixie•about 1 hour ago

Yeah - and if they wanted some cheap PR, that's one way to get it.

janalsncm•about 2 hours ago

Maybe another way to say it is to reframe the idea of “human in the loop”.

Humans are always in the loop, because we can always expand the definition of loop to include the humans that pushed the button and built the system and processes that happen after the button was pushed, and humans that ordered others to push the button. The level of direct involvement varies, but culpability doesn’t.

verdverm•about 2 hours ago

> It's pathfinding through language generation.

What an interesting sentence (to describe inference time reasoning)

tantalor•about 4 hours ago

Asinine.

"rogue" and "off leash" mean the same thing, the thing is not under control

Cyan488•about 3 hours ago

I think we should introduce "rampant" in the real-world AI vernacular

temp0826•about 3 hours ago

Pfhor what reason?

1659447091•about 3 hours ago

> "rogue" and "off leash" mean the same thing, the thing is not under control

To go "rogue" is to go against the control

To be "off leash" is to not be controlled

By releasing the automation, as the article says, without controls*, is what makes makes it "off leash" and not gone "rogue"

*"OpenAI gave the models a task with no answer, and no way to quit."

Models Don't Go Rogue

⚡ Community Insights

Discussion (16 Comments)Read Original on HackerNews