Raw LLM Responses
Inspect the exact model output for any coded comment.
Look up by comment ID
Random samples — click to inspect
G
The government should seize the land that was seized by white settlers all those…
rdc_deuinga
G
@structures77 its not my government (I'm a Brit bty) its all governments. Neithe…
ytr_UgxyOOBEV…
G
I think this news demostrates a growing trend to scapegoat AI for poor decision …
ytc_UgxOp5fL5…
G
AI cannot be and is not self-aware and what is more cannot have any form of emot…
ytc_UgxTH6Vr1…
G
What if Chinese AI and US AI collaborate to give those idiots advice that messes…
ytc_UgxCkty5T…
G
Try to imagine a world that doesn’t run on money. Quantum computing already had …
ytr_UgwesRSEY…
G
This society is cooked stop fucking with these robots it's like they want them t…
ytc_Ugz1UPmCF…
G
People just don't respond to any AI communication will be a start to shut it dow…
ytr_UgwbMGmrX…
Comment
At present I am running with the assumption that this is *NOT* a hacked server on the part of teamviewer (there's no evidence to support that assertion, yet), but is instead an exploitation of device discovery and "allow remote control". Allow remote control makes it so your computer is able to be remote controlled by a "support" technician, which is all well and good. Older versions of TeamViewer used 4 digit numbers (newer versions of Team viewer use a 6-character alphanumeric (a-z+0-9) randomly generated password.
With ~200,000,000 accounts and a 60 second cooldown after 4 guesses the attackers could easily just be brute forcing machines that allow for remote control, and sending a success message back to the program controlling it when they gain access. With a pool that large and discoverability / password guessing being easily brute forced (discovery comes via different messages if a computer doesn't exist or incorrect credentials), it seems that is a more likely scenario than a central server breach.
That said if this is the case TV is braindead for not catching the amount of incoming requests trying to brute force much sooner.
Note: This kind of access can be disabled -- http://i.imgur.com/ITX4ToC.png
If this proves to be a server breach I'll have to put in some serious hours this weekend to remove it. Ick.
Edit: I stated in another comment:
>It's also possible that this coincides with a larger hacking. The un/pw's were released from the linkedin hacking from years ago, and apparently just started to circulate (HaveIBeenPwned alerted me to it a few weeks ago). Thus password reuse could actually be the root cause here, despite my earlier claim. Email / Password combos could be automated and sent as well.
Which seems to coincide with the timing off these attacks as well. You can check to see if your credentials were leaked in this (or any other) attack at https://haveibeenpwned.com/ . If you reuse your passwords and did so for TeamViewe
reddit
AI Responsibility
1464916238.0
♥ 55
Coding Result
| Dimension | Value |
|---|---|
| Responsibility | none |
| Reasoning | unclear |
| Policy | unclear |
| Emotion | indifference |
| Coded at | 2026-04-25T08:33:43.502452 |
Raw LLM Response
[
{"id":"rdc_d3tvm39","responsibility":"none","reasoning":"unclear","policy":"unclear","emotion":"indifference"},
{"id":"rdc_d3u2xvy","responsibility":"none","reasoning":"unclear","policy":"unclear","emotion":"indifference"},
{"id":"rdc_nma9cv9","responsibility":"none","reasoning":"unclear","policy":"unclear","emotion":"mixed"},
{"id":"rdc_nm8zldl","responsibility":"company","reasoning":"consequentialist","policy":"unclear","emotion":"approval"},
{"id":"rdc_nma96nu","responsibility":"none","reasoning":"unclear","policy":"unclear","emotion":"mixed"}
]