Raw LLM Responses
Inspect the exact model output for any coded comment.
Look up by comment ID
Random samples — click to inspect
G
They know AI is a danger to our future and yet they do it anyway. No different t…
ytc_UgyHfsJTB…
G
I think the issue is if ai replaces humans in jobs who pays takes to governments…
ytc_Ugy-SyWvz…
G
not going to happen while i am a live nor after my since any generative AI comin…
ytc_UgwwnwYLu…
G
We're talking about the automation of human intellectual labour, whatever new jo…
ytc_Ugy_mUmaA…
G
AI might not be art, but I do think intentionally poisoning AI models has a pote…
ytc_Ugy4RJWGK…
G
I don't need emojis or friendship from chatGPT. I just want it to respond in a r…
rdc_njh2otg
G
The bananna, is a great example, or the buckets of sand falling over. Or the "ta…
ytc_UgxQihxxf…
G
AI is so inaccurate that we still need people to vet the AI responses for errors…
ytc_UgzjVxPZF…
Comment
I love Stephen Fry, but a lot of these predictions are beyond speculative, especially where they meet the physical world.
Just one example: if GPT-6 becomes dangerous, it's always possible to pull the power of the datacenters, or even bomb them. Rogue AIs 3d printing armies of killer drones? Again, destroy the plant.
This video distracts from the real dangers, which I think are threefold.
First, there is likely to be mass unemployment. I've been predicting this for more than 25 years. As our systems increasingly become scalable, fewer and fewer people are necessary to run them.
Second, military technology will become even more dangerous. I think that the war in Ukraine is the breeding ground for this. Once there are autonomous killer drones in the world it's a simple step for them to be used by groups like the Iranian Revolutionary Guard to carry out attacks.
Third, similar groups will try to use open source models to create bio-weapons. A friend who is a PhD CEO in pharmaceuticals says that it's much harder to create and deploy such weapons than we think, but I still expect it to happen.
youtube
AI Moral Status
2025-07-12T17:4…
Coding Result
| Dimension | Value |
|---|---|
| Responsibility | developer |
| Reasoning | consequentialist |
| Policy | regulate |
| Emotion | mixed |
| Coded at | 2026-04-27T06:24:59.937377 |
Raw LLM Response
[
{"id":"ytc_Ugzzmk3qv60b9rhhJDR4AaABAg","responsibility":"company","reasoning":"consequentialist","policy":"liability","emotion":"outrage"},
{"id":"ytc_Ugw8cJFu53aVrXzeO8p4AaABAg","responsibility":"unclear","reasoning":"unclear","policy":"unclear","emotion":"mixed"},
{"id":"ytc_Ugy6rhk3pkhsiCFA3CB4AaABAg","responsibility":"distributed","reasoning":"consequentialist","policy":"regulate","emotion":"outrage"},
{"id":"ytc_UgwFXRKKE8gQ2u_kvWV4AaABAg","responsibility":"company","reasoning":"consequentialist","policy":"liability","emotion":"mixed"},
{"id":"ytc_UgzKPEMgBhpFKK5xmgx4AaABAg","responsibility":"ai_itself","reasoning":"mixed","policy":"none","emotion":"approval"},
{"id":"ytc_UgwEKvuU7r0hTku-mQN4AaABAg","responsibility":"none","reasoning":"deontological","policy":"none","emotion":"fear"},
{"id":"ytc_UgwVicA4rp3Yf49154N4AaABAg","responsibility":"distributed","reasoning":"consequentialist","policy":"none","emotion":"fear"},
{"id":"ytc_UgxoDt1txHnUatRQccF4AaABAg","responsibility":"developer","reasoning":"consequentialist","policy":"regulate","emotion":"mixed"},
{"id":"ytc_UgyBbam9hcw7YIqG_3t4AaABAg","responsibility":"user","reasoning":"deontological","policy":"liability","emotion":"fear"},
{"id":"ytc_Ugyl7DvKAaXllbxYeUh4AaABAg","responsibility":"ai_itself","reasoning":"unclear","policy":"none","emotion":"mixed"}
]