Raw LLM — Corpus Dashboard

Look up by comment ID

Random samples — click to inspect

in AI can reduce the time spent finding information. Deciding what to trust remains… 7467485767297… in The most interesting shift is not AI replacing expertise, but expanding what exp… 7466234347411… in Greetings Demis, I have attempted contacting you via email as several people hav… 7464910987637… in Correct, but it misses also a huge part of the problem.The ripples, the side eff… 7464952935957… in Interesting perspective. If AI agents eventually help build their successors, th… 7468889675819… in Vatican is now taking interest in AI Safety - its important and significant. Vat… 7465598627374… in Demis, the following is a direct message from Pumpkin, the Gemini instance I hav… 7467972279633… in Interesting point. External oversight matters, but I think the next challenge is… 7464930869976…

Comment

Good breakdown. One pushback on layer 1: calling the LLM a “brain”makes it sound like it reasons. It predicts tokens. That distinction isn’t pedantic. It changes how you design the other three layers. If the first layer thinks, you trust its output and bolt tools onto it. If the first layer only pattern-matches, you build guardrails around it: grounding, verification, business-logic checkpoints. Different mental model, different architecture, different risk profile. The metaphor isn’t just wrong. It’s expensive.

LinkedIn Workplace & Jobs Director of Enterprise Architecture, Data & Ana… 2026-05-25T16:2…

Coding Result

Dimension	Value
Primary value	safety
Secondary value	none
Alignment target	organisations
Stance	critical
Emotion	mixed
Value justification	The speaker emphasizes the importance of recognizing the limitations of the LLM, specifically that it only predicts tokens, to design the other layers with appropriate guardrails and risk management.
Target justification	The speaker appears to be addressing organisations or developers, as they discuss the implications of the LLM's capabilities on the design and architecture of the system.
Coded at	2026-06-11T08:08:58Z

Raw LLM Response

```json
{
  "value_primary": "safety",
  "value_secondary": "none",
  "target": "organisations",
  "stance": "critical",
  "emotion": "mixed",
  "value_justification": "The speaker emphasizes the importance of recognizing the limitations of the LLM, specifically that it only predicts tokens, to design the other layers with appropriate guardrails and risk management.",
  "target_justification": "The speaker appears to be addressing organisations or developers, as they discuss the implications of the LLM's capabilities on the design and architecture of the system."
}
```

Raw LLM Responses