AI ALIGNMENT FORUM
AF

Wikitags

Tool AI

Edited by Alex_Altair, Multicore, et al. last updated 30th Dec 2024

Tool AI is a type of Artificial Intelligence that is built to be used as a tool by the creators, rather than being an agent with its own action and goal-seeking behavior.

Generally meant to refer to AGI, tool AI is a proposed method for gaining some of the benefits of the intelligence while avoiding the dangers of having it act autonomously. It was coined by Holden Karnofsky, co-founder of GiveWell, in a critique of the Singularity Institute. Karnofsky proposed that, while he agreed that agent-based AGI was dangerous, it was an unnecessary path of development. His example of tool AI behavior was Google Maps, which uses complex algorithms and data to plot a route, but presents these results to the user instead of driving the user itself.

Eliezer Yudkowsky responded to this by enumerating several ways in which tool AI had similar difficulties in technical specification and safety. He also pointed out that it was not a common proposal among leading AGI thinkers.

See Also

  • Oracle AI

External Links

  • Conversation between Holden Karnofsky and Jaan Tallinn
Subscribe
1
Subscribe
1
Discussion0
Discussion0
Posts tagged Tool AI
28Solving the whole AGI control problem, version 0.0001
Steven Byrnes
4y
2
26What is a Tool?
johnswentworth, David Lorell
1y
0
15[Intro to brain-like-AGI safety] 11. Safety ≠ alignment (but they’re close!)
Steven Byrnes
3y
0
141Simulators
janus
3y
90
72Cyborgism
NicholasKees, janus
3y
20
42The Fusion Power Generator Scenario
johnswentworth
5y
10
60Interpretability/Tool-ness/Alignment/Corrigibility are not Composable
johnswentworth
3y
4
44Applying superintelligence without collusion
Eric Drexler
3y
19
16A multi-disciplinary view on AI safety research
Roman Leventov
3y
0
14Some reasons why a predictor wants to be a consequentialist
Lauro Langosco
3y
0
6Why not tool AI?
Q
smithee, Ben Pace
7y
Q
2
Add Posts