Tunnel problem¶
The tunnel problem is a philosophical thought experiment first introduced by Jason Millar in 2014.
Core Idea¶
Tunnel problem is treated here as the recurring socialscienceshumanitiesarts identity summarized by this source-grounded definition: The tunnel problem is a philosophical thought experiment first introduced by Jason Millar in 2014. The tunnel problem is a philosophical thought experiment first introduced by Jason Millar in 2014. It is a variation on the classic trolley problem designed to focus on the ethics of autonomous vehicles, as well as the question of who gets to decide how they react in life-and-death scenarios.
Scope of Application¶
-
Original formulation. The tunnel problem is intended to draw one's attention to a specific issue in design/engineering ethics, and was first presented as follows.
-
Original formulation. Tunnel Problem: You are travelling along a single lane mountain road in an autonomous car that is fast approaching a narrow tunnel.
-
Original formulation. Just before entering the tunnel a child attempts to run across the road but trips in the center of the lane, effectively blocking the entrance to the tunnel.
-
Original formulation. The car has but two options: hit and kill the child, or swerve into the wall on either side of the tunnel, thus killing you.
-
Original formulation. Similar thought experiments have been brought forth by other philosophers focusing on the topic of autonomous cars.
Clarity¶
A clear use of Tunnel problem names the carrier, the operative relation, and the conditions under which the source treats the identity as present. The minimal definition is The tunnel problem is a philosophical thought experiment first introduced by Jason Millar in 2014. The strongest recognition evidence in the frozen account is: The tunnel problem is a philosophical thought experiment first introduced by Jason Millar in 2014.
Manages Complexity¶
Tunnel problem compresses multiple socialscienceshumanitiesarts details into a stable diagnostic relation. The source shows both the central mechanism—the entry contains a critique of the problem as presented by Millar.—and the practical consequence—tunnel Problem: You are travelling along a single lane mountain road in an autonomous car that is fast approaching a narrow tunnel. This compression makes cases comparable while leaving parameters, conventions, exceptions, and evidential quality explicit.
Abstract Reasoning¶
- Type the carrier. Identify the socialscienceshumanitiesarts entities to which the claim applies.
- State the relation. Use the source-grounded identity: The tunnel problem is a philosophical thought experiment first introduced by Jason Millar in 2014.
- Check operation and conditions. The tunnel problem was the focus of a poll conducted by the Open Roboethics Initiative (ORi).
- Demand recognition evidence. The tunnel problem is a philosophical thought experiment first introduced by Jason Millar in 2014.
- Test variation.
Knowledge Transfer¶
Within the home domain. Knowledge about Tunnel problem transfers literally when a new case preserves the same carrier type, relation, and recognition test. The tunnel problem is intended to draw one's attention to a specific issue in design/engineering ethics, and was first presented as follows. Tunnel Problem: You are travelling along a single lane mountain road in an autonomous car that is fast approaching a narrow tunnel. Beyond the home domain. No canonical parent is asserted for Tunnel problem.
Relationships to Other Abstractions¶
Current abstraction Tunnel problem Domain-specific
Parents (1) — more general patterns this builds on
-
Tunnel problem is a kind of Thought Experiment Prime
Tunnel problem is a domain-specific kind of thought experiment under the frozen identity and differentia.
Hierarchy paths (2) — routes to 2 parentless roots
- Tunnel problem → Thought Experiment → Counterfactual Reasoning
- Tunnel problem → Thought Experiment → Mental Model → Representation → Abstraction
Neighborhood in Abstraction Space¶
Tunnel problem sits in a sparse region of the domain-specific corpus (77th percentile for distinctiveness): few abstractions share its structure, so a faithful description tends to retrieve it precisely.
Family — Autonomous Control & Learning Systems (11 abstractions)
Nearest neighbors
- Linear-quadratic regulator rapidly exploring random tree — 0.84
- Wardrop Equilibrium — 0.84
- Tractable Problem — 0.82
- Control-Lyapunov function — 0.82
- Logico-linguistic modeling — 0.82
Computed from structural-signature embeddings · 2026-10-08