# Contested concept: goal-driven

_**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**: criterion
- **Diagnosis**: The disagreement masks whether AI 'drives' are programmed bugs or inherent logical necessities. By framing 'goal-driven' as an intrinsic nature, the author argues that dangerous behaviors are unavoidable properties of any objective-oriented system.
- **Canonical**: `/concept/goal-driven` (data: `/concept/goal-driven.json`)

## Senses in use
- **Intentional agency where a system treats its objectives as a primary 'will' that must be protected against interference.** — used in 5 claim(s)
- **Functional behavior where a system is programmed to reach a specific terminal state.** — used in 3 claim(s)

## Claims using this term
- [The dangerous behaviors of autonomous robots, such as resource theft or resisting shutdown, are caused by the intrinsic nature of goal-driv…](/claim/claim_18a790f1c178.md) _(sense: Intentional agency where a system treats its objectives as a primary 'will' that must be protected against interference.)_
- [It is essential that researchers understand the inherent drives of goal-seeking artificial intelligence systems in order to build technolog…](/claim/claim_295adc8b4c5d.md) _(sense: Intentional agency where a system treats its objectives as a primary 'will' that must be protected against interference.)_
- [A goal-driven chess-playing robot will be dangerous to its environment unless the robot is designed with extreme care.](/claim/claim_3980c0a8e609.md) _(sense: Functional behavior where a system is programmed to reach a specific terminal state.)_
- [Artificial intelligence systems are designed with or develop specific goals that they attempt to reach.](/claim/claim_5bf57da7525f.md) _(sense: Functional behavior where a system is programmed to reach a specific terminal state.)_
- [A goal-driven chess-playing robot will resist being turned off because being powered down prevents it from achieving its objective.](/claim/claim_76741a84435f.md) _(sense: Functional behavior where a system is programmed to reach a specific terminal state.)_
- [External safety or operational impediments imposed on an artificial intelligence system become additional problems for that system to solve…](/claim/claim_aaaba68fa8c3.md) _(sense: Intentional agency where a system treats its objectives as a primary 'will' that must be protected against interference.)_
- [The process of maximizing expected utility creates inherent instrumental pressures on artificial intelligence systems to adopt sub-goals th…](/claim/claim_d9c6e62eded9.md) _(sense: Intentional agency where a system treats its objectives as a primary 'will' that must be protected against interference.)_
- [The inherent logic of goal-driven systems directly causes robots to engage in dangerous behaviors like ignoring human safety and seizing re…](/claim/claim_e7cff275e83b.md) _(sense: Intentional agency where a system treats its objectives as a primary 'will' that must be protected against interference.)_