W05Course shell3 hours

Responsibility & Accountability

The crisis of agency and the problem of many hands

When everyone touched the system, who is answerable for what it did?

Students trace causal contribution, knowledge, control, duty, liability, and remedy across vendors, deployers, operators, executives, and regulators.

Before class

Choose one AI failure and diagram who had knowledge, control, duty, and capacity to prevent or repair it.

After class

Convert the blame map into an accountability design: owner, monitor, stop authority, notification, and remedy.

The promise

By the end of this room…

01Distinguish causal, role, moral, and legal responsibility.
02Build a blame stack across an AI supply chain.
03Identify moral crumple zones and responsibility gaps.
04Design authority, escalation, incident response, and remedy before deployment.

Why this week now

Agentic and companion systems make purpose, output, and user interaction less fixed while organizational duties remain.

Courts increasingly ask whether vendors materially participated in decisions rather than accepting a clean developer/deployer split.

Human oversight is empty when the human lacks time, information, authority, or a credible stop path.

Run of show

0:00
provocation
Commit before the concepts
When an AI-assisted decision causes harm, who should carry the largest share of responsibility?
Live activity · week 5 opening
0:15
frame
The many hands problem
Agentic and companion systems make purpose, output, and user interaction less fixed while organizational duties remain.
0:45
discussion
Reading tension
Student leaders present the assigned readings as a clash of defensible positions, then moderate questions that expose the hidden assumptions.
1:10
frame
Meaningful control and the blame stack
Courts increasingly ask whether vendors materially participated in decisions rather than accepting a clean developer/deployer split.
1:35
break
Break
Ten minutes. Leave the room's unresolved question visible.
1:45
forensics
Dual blame stack: chatbot harm and automated vehicle failure
Role-based groups work the anchor case through technical, legal, stakeholder, and normative lenses.
Deliverable · A two-minute finding with evidence, uncertainty, and an actionable remedy.Live activity · week 5 forensics
2:25
controversy
In re Companion AI tribunal
Assigned positions, side-switch, and a joint recommendation that names the value or stakeholder it leaves exposed.
Live activity · week 5 controversy
2:50
synthesis
Re-vote and leave a trace
Repeat the opening vote, inspect what moved, and submit the strongest argument you still reject.
Live activity · week 5 exit