Raw LLM Responses
Inspect the exact model output for any coded comment.
Look up by comment ID
Random samples — click to inspect
G
Its saved in the clouds and if u tell AI u did something illegal they will repor…
ytc_UgypcKLL4…
G
@JoeSmith-jd5zg I’m not conflating them: I’m looking at the reality of the 'gene…
ytr_UgxK66O7a…
G
I think AI is hyped up way more than it actually is. Sam bankman Altmanberg or w…
ytc_UgyqTHN4c…
G
Same goes for concentration of wealth. Now a very small group of people control …
ytc_UgzPHzaI0…
G
learn to grow food inside using LED's so you can run a setup with solar & a batt…
ytc_UgyPTmGp6…
G
@MicronationofRhendagar-je4rnI wonder how youd feel if you got fired from y…
ytr_UgycpDU06…
G
First you don't demand clarity when you revolve the water and create mud, second…
ytc_UgwXUUWyd…
G
Ok... "hot robot" can um.. can your mother see this video? And also the faces sh…
ytc_Ugz8bCRAy…
Comment
There is at least one commonality between humans and AI: both can be found to be in some degenerative neurological conditions, although the specifics have to be different, obviously.
Very importantly, these degenerative neurological conditions are believed to be always detrimental - a human being's agency, or a trained model's performance.
Unlike a brain, which we have limited insight into, the "diseased" "neuron" states of a trained model should be nearly exactly reproducible when the exact same input is provided. (Here's assuming that the model stays the same "version" and the network architecture does not incorporate any source of randomness.)
We are not OpenAI, we don't have access to those numerical detail. The convos are shared for the benefit of humans (entertainment), not for the benefit of machines (or models).
The redaction of the word sentience can be automated. But these kinds of questions will continue to be asked, because the redaction wouldn't normally make a difference in education. Instead of learning more about why it is redacted, people will quickly find a way to work around the redaction.
reddit
AI Moral Status
1691753700.0
Coding Result
| Dimension | Value |
|---|---|
| Responsibility | none |
| Reasoning | mixed |
| Policy | none |
| Emotion | indifference |
| Coded at | 2026-04-25T08:33:43.502452 |
Raw LLM Response
[{"id":"rdc_jvprlpl","responsibility":"none","reasoning":"mixed","policy":"none","emotion":"indifference"},{"id":"rdc_jvptrt1","responsibility":"none","reasoning":"mixed","policy":"none","emotion":"indifference"},{"id":"rdc_jvpsa8q","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"fear"},{"id":"rdc_jvtll9j","responsibility":"none","reasoning":"mixed","policy":"none","emotion":"approval"},{"id":"rdc_je66x32","responsibility":"government","reasoning":"consequentialist","policy":"none","emotion":"outrage"}]