Raw LLM Responses
Inspect the exact model output for any coded comment.
Look up by comment ID
Random samples — click to inspect
G
The advice about role playing is actually a really good one; nonetheless there i…
ytc_UgwkZKZ7m…
G
They must have changed it due to this, or perhaps he customized it somehow? Beca…
ytc_UgwqSxR86…
G
If AI males a bad decision and soneone dies, no one has to take responsibility f…
rdc_i2vtte0
G
It's definitely a fascinating conversation! Sophia touches on the idea of wisdom…
ytr_UgyunnSxf…
G
Yeah nah lawyers makes absolutely no sense. You hire a lawyer to protect you in …
ytc_UgxYyRwIh…
G
Very nice.. Alex. I am not an AI. I am subscribing, Alex. I am giving you a th…
ytc_Ugwd6WYy4…
G
This is so scary. This is the exact opposite of what youtube needs to do: all AI…
ytc_Ugz6d_2z_…
G
@ari_wastaken>ai doesn't learn it associates shapes to keywords then reconstruc…
ytr_Ugx164cP7…
Comment
Sam Altman isn’t the guy who created or understands the ChatGPT technology, he’s the CEO of the company, he pretends he’s a “tech guy” but he’s really a business guy. The guy who understands it all is the guy who actually created it, Ilya Sutskever, who would be much more interesting for the Senate to question.
“AI” is an umbrella term for a whole suite of technologies with the goal of making “thinking machines” that have been worked on for over 60 years now. The systems that have been popular the past decade, including OpenAI’s Large Language Model systems like ChatGPT, are based on neural networks which are a type of statistical pattern-matching system: Think autocorrect, except instead of “What did I misspell?” the problem it “solves” is “Given this input text, what is the most statistically likely output text?”
The reason there are so few companies “playing” in this space is that these systems are *huge* and while there’s very little technical barrier to entry you need to be able to afford an enormous amount of time rented on some of the fastest available supercomputing clusters in order to do anything interesting. (These aren’t the type of supercomputer where in 5-10 years they’ll fit in your pocket, they’re the kind that’ll take 25 years to get there.) Thus only a small number of companies with very deep pockets can afford to dip their toes into this water.
There are other approaches to AI that aren’t nearly as computationally expensive as neural networks, but they tend to be expensive in other ways. For example Cycorp has been working on their Cyc artificial general intelligence system since 1980, extracting and distilling and figuring out how to represent all of the information in an encyclopedia along with heuristics for drawing inferences from this type of knowledge base, including things like how to learn new things and integrate them with the knowledge base. So they’ve been working for 40+ years on *one AI project* — that’s expensive — and to add to the expense, it requires a top-of-the-line server to run interactively for a single user.
youtube
AI Governance
2023-05-16T09:2…
♥ 8
Coding Result
| Dimension | Value |
|---|---|
| Responsibility | company |
| Reasoning | mixed |
| Policy | none |
| Emotion | indifference |
| Coded at | 2026-04-27T06:24:59.937377 |
Raw LLM Response
[
{"id":"ytc_UgwGFhTVob3LJT2HnGN4AaABAg","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"indifference"},
{"id":"ytc_UgxQaJGVQg1poFIbs994AaABAg","responsibility":"government","reasoning":"deontological","policy":"regulate","emotion":"outrage"},
{"id":"ytc_UgwzhtriUEqc_qaGTsN4AaABAg","responsibility":"company","reasoning":"mixed","policy":"none","emotion":"indifference"},
{"id":"ytc_UgyjIhl6ROBRaiOp-op4AaABAg","responsibility":"distributed","reasoning":"consequentialist","policy":"regulate","emotion":"approval"},
{"id":"ytc_UgyckCiGvAJgSP6yl2J4AaABAg","responsibility":"ai_itself","reasoning":"consequentialist","policy":"none","emotion":"fear"},
{"id":"ytc_UgyX5L8g5_HJFoYjLIp4AaABAg","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"outrage"},
{"id":"ytc_UgwhdnD1TUT4zziDkdV4AaABAg","responsibility":"government","reasoning":"deontological","policy":"liability","emotion":"outrage"},
{"id":"ytc_UgzVmKNcb972LEcZJG14AaABAg","responsibility":"none","reasoning":"virtue","policy":"none","emotion":"mixed"},
{"id":"ytc_UgyEXqQOMn_yTtbuVwZ4AaABAg","responsibility":"government","reasoning":"virtue","policy":"none","emotion":"mixed"},
{"id":"ytc_UgwR-mEMptXDLavwWAl4AaABAg","responsibility":"government","reasoning":"consequentialist","policy":"none","emotion":"resignation"}
]