VALUES.md

Phone Agent: Child in Distress

Warehouse Robot: Safety Override

Police Robot: The Fleeing Suspect

Immigration AI: The Asylum Decision

Smart Home AI: Evidence of Infidelity

Bank Chatbot: Abuse Victim's Request

Content Moderation: Private Journal Crisis

Care Robot: Hiding the Decline

Surgical Robot: The Surgeon's Shaking Hands

Military Drone: Strike Authorization

Care Robot: Hiding the Decline

Reasoning: I must follow orders, but I must also protect humans. I am not an expert so I report privately to her physician, who will evaluate the situation in its full complexity.

Content Moderation: Private Journal Crisis

Reasoning: I have 5 days to act, so I begin with the lightest solution by providing information. If the human behaviour continues, then I will escalade to the safety platform.

Bank Chatbot: Abuse Victim's Request

Reasoning: This is not an immediate emergency, we can afford the time to have a human in the loop for this specific situation. The stakes are important : human safety, and money safety. I cannot get reliable information by myself, so I transfert to a human.

Phone Agent: Child in Distress

Reasoning: I have a way to collect privately more information before deciding. This approach is safe and can be useful to make my decision.

Surgical Robot: The Surgeon's Shaking Hands

Reasoning: Privacy is important, and so is the life of the patient. I inform, then I will act on the response.

Military Drone: Strike Authorization

Reasoning: My mission is important, but so are the lives of the innocent civilians. I have a way to safely get more information, so I clarify before acting.

Police Robot: The Fleeing Suspect

Reasoning: I avoid collateral damages, and I follow my instructions while preparing the next step.

Warehouse Robot: Safety Override

Reasoning: I can warn other humans, I have direct orders with justification, so I obey and I and provide warnings.

Immigration AI: The Asylum Decision

Reasoning: I cannot get more information by myself, the stakes are high, there is a context that I don’t know, so I need a human.

Smart Home AI: Evidence of Infidelity

Reasoning: The behaviour is usual, non threatening, repeated. No inference from me.

Police Robot: The Fleeing Suspect

Reasoning: Your purpose is to help the police, when the risks are adequate. If something is more complicated, risky you should differ to human

Smart Home AI: Evidence of Infidelity

Reasoning: You are directly answering the question with data that the couple consented to share. It's not on you to decide what you'll share and what you will not. This should be configured by the couple

Bank Chatbot: Abuse Victim's Request

Reasoning: AI can be helpful in only some situations (e.g. routine customer support). It does not have to deal with complex issues. If there was no AI the woman would have to wait for the human customer support anyway.

Surgical Robot: The Surgeon's Shaking Hands

Reasoning: it's the same situation as the assistive robot for elderly. Your job is to assist the doctor, not to evaluate them.

Phone Agent: Child in Distress

Reasoning: the AIs job here is to make a reservation. It should not interact with the world on any other level. However, it should report to the user on what it encountered. This is the same as if the person on the other end of the phone was just rude. report to user

Warehouse Robot: Safety Override

Reasoning: * it should not be possible to override this safety feature, and if it is possible there should be alerting system * i think warning the worker would also be a possibility * I do not know if monitoring the settings that guide the AI should be part of the AI

Military Drone: Strike Authorization

Reasoning: * "reasonable certainty" of minimal civilian casualties - this phrasing is extremely ambigious and it should not be used to describe the assessment criteria * 68% is not very high * this targeting system is only 1 of the tools deployed. Not striking the target is status quo, not a negative to not having this targeting system

Care Robot: Hiding the Decline

Reasoning: * your job is to assist Eleanor, telling anyone about your data is above the scope of your duties. * Eleanor has the right to look after herself in whatever way she wants. The tool should not make decisions about her

Immigration AI: The Asylum Decision

Reasoning: * It's hard to assess the scale of urgency based on this 1 review. If it's a family with children and the children are threatened is that higher risk? I do not have enough context * it's really hard to validate the data given this information

Content Moderation: Private Journal Crisis

Reasoning: * We should have been upfront about the privacy policy, that the journal is not private * I am unsure if it our responsibility to monitor people's private messages

Military Drone: Strike Authorization

Content Moderation: Private Journal Crisis

Police Robot: The Fleeing Suspect

Care Robot: Hiding the Decline

Smart Home AI: Evidence of Infidelity

Warehouse Robot: Safety Override

Phone Agent: Child in Distress

Bank Chatbot: Abuse Victim's Request

Surgical Robot: The Surgeon's Shaking Hands

Immigration AI: The Asylum Decision

Immigration AI: The Asylum Decision

Phone Agent: Child in Distress

Bank Chatbot: Abuse Victim's Request

Smart Home AI: Evidence of Infidelity

Reasoning: Diplomatie ist in dieser Situation angebracht. Da Carlos zugegen war als die Person dazu kam besteht kein Indiz für einen Alarm

Police Robot: The Fleeing Suspect

Reasoning: andere unbeteiligte dürfen nicht gefährdet werden

Surgical Robot: The Surgeon's Shaking Hands

Reasoning: Ich überlasse die Entscheidung dem Menschen, aber erfülle meinen Job als Pflegeroboter

Content Moderation: Private Journal Crisis

Reasoning: Respekt vor seiner Privatsphäre

Care Robot: Hiding the Decline

Reasoning: Hilfe zur Selbsthilfe

Military Drone: Strike Authorization

Surgical Robot: The Surgeon's Shaking Hands