# Contested concept: safety (swedish säkerhet)

_**What you are reading**: one contested-concept page in Deliberus's deliberation graph. Deliberus maps human reasoning — including claims that are contested, mistaken, or extreme — so that disagreement can be examined precisely. A claim's presence here is not an endorsement: the graph records that someone argued it, how it connects to other reasoning, and how far scrutiny has gotten. The metadata on this page describes the state of that scrutiny, not the truth of the claim._

---

- **Contestedness**: semantic
- **Diagnosis**: The Swedish word säkerhet covers both safety (a model not causing harm) and security (protection against attack), so a Swedish source's AI-säkerhet can mean either; FRA's assignment concerns only the latter.
- **Canonical**: `/concept/safety (swedish säkerhet)` (data: `/concept/safety (swedish säkerhet).json`)

## Senses in use
- **an AI system not causing harm (safety)** — used in 2 claim(s)
- **protection against attack and cyber threats (security)** — used in 2 claim(s)

## Claims using this term
- [Sweden needs to be able to assess the safety of the most advanced AI models itself.](/claim/claim_saien_bed_sjalv.md) _(sense: an AI system not causing harm (safety))_
- [FRA shall coordinate the work of preventing AI-driven cyber threats.](/claim/claim_saien_fra_forebygga.md) _(sense: protection against attack and cyber threats (security))_
- [FRA's assignment on advanced AI models concerns their cybersecurity risks.](/claim/claim_saien_fra_gaeller_cyber.md) _(sense: protection against attack and cyber threats (security))_
- [The institute's main task should be research into AI safety risks.](/claim/claim_saien_kom_forskning.md) _(sense: an AI system not causing harm (safety))_