Raw LLM — Corpus Dashboard

Look up by comment ID

Random samples — click to inspect

in Right, weakest layer often determines success of AI stack, especially integratio… 7464632825892… in ATTENTION: @Demis Hassabis & the Google DeepMind Safety Architecture Team Consid… 7465483074831… in You forgot to add Hyperlambda.dev The Best Solution for building AI Agents with … 7462284159554… in Excellent post pro From a security perspective: LLM: Protect against prompt inje… 7464722787039… in This breakdown is excellent, Luís. What I see in real systems is that the “body”… 7464950955843… in Scott Schobert Totally agree! This is a major concern that is surfacing right no… 7468743178913… in This is a very important shift. The real value in learning AI is not collecting … 7466277292600… in IMO, Magnifica Humanitas is not simply about “AI ethics” in the usual sense. It … 7465242799047…

Comment

Sabrina N. I don't think there is any one solution. I don't think bans and the use of AI detection are the way to go. They are not meaningfully enforceable or fit for purpose, respectively. It's clear at this stage that assessment needs to change. The days of relying on artefacts as stand-ins for learning are probably over (and that has been well overdue for some time, as someone who has been investigating contract cheating for years). If for some reason a university wants to use a report or an essay for the purposes of assessment, they can either, a) use it as a purely formative exercise (remove the value of cheating), b) watch the student write it, or c) make the assessment a face-to-face conversation about the document rather than the document itself. Ultimately, universities need to be spending more time having conversations with students about their learning, and these conversations should be the assessment.

LinkedIn General AI Discourse Integrity Investigator & Data Scientist | Assur… 2026-05-28T04:2…

Coding Result

Dimension	Value
Primary value	human_autonomy
Secondary value	none
Alignment target	individual_users
Stance	demanding
Emotion	approval
Value justification	The speaker emphasizes the need for universities to focus on conversations with students about their learning, implying a desire for human judgment and autonomy in the assessment process.
Target justification	The speaker is primarily concerned with the impact of AI on individual students and their learning experience, as evident from their suggestions for alternative assessment methods.
Coded at	2026-06-11T08:26:26Z

Raw LLM Response

```json
{
  "value_primary": "human_autonomy",
  "value_secondary": "none",
  "target": "individual_users",
  "stance": "demanding",
  "emotion": "approval",
  "value_justification": "The speaker emphasizes the need for universities to focus on conversations with students about their learning, implying a desire for human judgment and autonomy in the assessment process.",
  "target_justification": "The speaker is primarily concerned with the impact of AI on individual students and their learning experience, as evident from their suggestions for alternative assessment methods."
}
```

Raw LLM Responses