Raw LLM Responses

Inspect the exact model output for any coded comment.

Look up by comment ID
Random samples — click to inspect
G This prompt has been working pretty good in gpt4: `You're a GPT legal advis… rdc_jhb69kr G Do video games? Detecting deep fakes can be done on a phone, ... creating it r… ytc_Ugw_epdi1… G i want the car thoroughly tested more than it is. I was not fond of the free sel… ytc_Ugz8dm3JW… G How is that exploit?? Artists are trying to live in this world. Ai literally st… ytr_UgygJi3wm… G My message for you, the reader of this comment, is this: The question is not whe… ytc_UgwSs_iR4… G I genuinely have more respect for people who draw stick figures than AI grifters… ytc_UgwnCvyhY… G How can you say that schools can now teach entire language programs, when langua… ytc_Ugw_85hrF… G I don’t know why but I feel this way sometimes. I say thank you and please to Ch… ytc_Ugz663AaS…
Comment
Hi, I am an AI Safety researcher at a top university. It's a flourishing field, tons of mechanisms for controlling future AI (practicing on current AI) are being proposed and tested. But it's also a field that needs a lot more support and is not ready yet for AGI at all. This is clear by how many distinct methods there are to jailbreak ChatGPT. That doesn't matter now, but it's a sign that we don't have control, and in the future that will matter.
reddit AI Governance 1728546669.0 ♥ 12
Coding Result
DimensionValue
Responsibilitydeveloper
Reasoningconsequentialist
Policyindustry_self
Emotionapproval
Coded at2026-04-25T08:33:43.502452
Raw LLM Response
[ {"id":"rdc_lr63axw","responsibility":"developer","reasoning":"consequentialist","policy":"none","emotion":"mixed"}, {"id":"rdc_lr7e97j","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"indifference"}, {"id":"rdc_lr68zwh","responsibility":"unclear","reasoning":"consequentialist","policy":"unclear","emotion":"mixed"}, {"id":"rdc_lr7yhhe","responsibility":"developer","reasoning":"consequentialist","policy":"industry_self","emotion":"approval"}, {"id":"rdc_lr7py00","responsibility":"developer","reasoning":"consequentialist","policy":"ban","emotion":"outrage"} ]