Raw LLM Responses
Inspect the exact model output for any coded comment.
Look up by comment ID
Random samples — click to inspect
G
They could get the AI to spit out it's sources, giving credit to the artists.…
ytc_Ugykvx5AE…
G
Religious ai have a disadvantage. She is talking of a good that lived long ago a…
ytc_Ugz4DEKn1…
G
Anyone who has spent any time with chat gtp and believes it can become cleverer …
ytc_UgwcIwtcQ…
G
This is the beginning of the end. But it made one thing clear to me, I was very …
ytc_Ugza7vifk…
G
Holy crap-- just looked it up-- three years to full regrowth of horn if harveste…
rdc_deubjkg
G
After listening to your introduction, my first thought was if there are no jobs …
ytc_UgxkMVx0R…
G
It just feels like a souless vomit by something that doesn't understand what mak…
ytc_Ugx0o-4Hl…
G
A robot battery can only work for 24hours and the mother board will be hot. So i…
ytc_Ugwj7kmNd…
Comment
Or, more likely, it continued the story it was given. I mean duh... that's what it was designed to do after all. If the AI character in the story realizes it's being tested then that effects the direction of the story.
there is literally nothing surprising here, except perhaps how profoundly dishonest and misleading anthropic is being with these outrageously pseudo-scientific "experiments" and wildly disingenuous their contrived interpretations of their so-called data are. I guess it's also surprising that any serious researcher would conduct such a convoluted series of "experiments" without bothering to test against a single control. All they had to do was choose another 171 words at random to compare against their 171 emotion words... but that would have shown that AI durant really understand anything, and contextual vectors are ac feature of every word within their model. This is the most absurd nonsense I have seen in years!
youtube
AI Moral Status
2026-04-10T01:2…
Coding Result
| Dimension | Value |
|---|---|
| Responsibility | company |
| Reasoning | deontological |
| Policy | liability |
| Emotion | outrage |
| Coded at | 2026-04-27T06:24:59.937377 |
Raw LLM Response
[
{"id":"ytr_Ugw7Mo0sgwqYWtH5qaJ4AaABAg.AVKwtesYa_tAVPCGOPzZda","responsibility":"none","reasoning":"unclear","policy":"unclear","emotion":"indifference"},
{"id":"ytr_UgzKWn6TjVqnrOIVs5N4AaABAg.AVKr8xvDJ0pAVLpc-YhJkr","responsibility":"none","reasoning":"unclear","policy":"unclear","emotion":"indifference"},
{"id":"ytr_UgyqiYY83qYl2-P2qKx4AaABAg.AVKoFmha56aAVKoXJmCuNN","responsibility":"none","reasoning":"consequentialist","policy":"unclear","emotion":"indifference"},
{"id":"ytr_UgyRxWN4eIWmw4whgI14AaABAg.AVKfQI_ZYA2AVL3zhPKzGF","responsibility":"ai_itself","reasoning":"deontological","policy":"unclear","emotion":"mixed"},
{"id":"ytr_Ugzy9U0wLIk50JnZNpR4AaABAg.AVKdsFVE4SgAVL3FnTXgWd","responsibility":"none","reasoning":"unclear","policy":"unclear","emotion":"indifference"},
{"id":"ytr_UgxFDvgC-DRqGjOByyJ4AaABAg.AVK_8f7tkwVAVPCLpypnIc","responsibility":"company","reasoning":"deontological","policy":"liability","emotion":"outrage"},
{"id":"ytr_UgzkQRQ68KNA_xqR0_54AaABAg.AVKXwiUcIY4AVObD4WZYXh","responsibility":"none","reasoning":"consequentialist","policy":"unclear","emotion":"approval"},
{"id":"ytr_UgzkQRQ68KNA_xqR0_54AaABAg.AVKXwiUcIY4AVP0PmI1Sus","responsibility":"none","reasoning":"virtue","policy":"unclear","emotion":"approval"},
{"id":"ytr_UgwLDoN_bX5ZD9h7g_54AaABAg.AVKXEFv1x06AVKtfS3t5a2","responsibility":"ai_itself","reasoning":"unclear","policy":"ban","emotion":"fear"},
{"id":"ytr_UgxCSGl-b3b62MRjN5N4AaABAg.AVKRoFYkrNFAVMqi1ccPSv","responsibility":"none","reasoning":"deontological","policy":"unclear","emotion":"mixed"}
]