Artificial intelligence and machine learning in armed conflict: a human-centred approach

Summary

The ICRC sets out how AI and machine learning bear on armed conflict and on humanitarian action. The paper looks at autonomy in weapons, cyber and information operations, and decision-making that affects lives. It argues that these systems should serve human actors and augment decision-makers rather than replace them, and that international humanitarian law still applies. It is cited here as a source and as an example of a human-centred public position. That citation does not mean the ICRC endorses, partners with, or is affiliated with AI Concerns.

The principles it rests on

In our words.

1. Human-centred use: systems serve people; they do not take their place.
2. Human control and judgment in decisions that affect life and dignity.
3. IHL remains the frame for AI used in or for conflict.
4. Humanitarian consequences are named before the tool is praised.

Scope

Use of AI and machine learning in the conduct of warfare or situations of violence, and in humanitarian action to assist and protect people affected by armed conflict.

Why it is exemplary

It is early, public, and written from protection rather than from product. It keeps human judgment in the sentence. It treats humanitarian action and the conduct of hostilities in the same frame.

Concerns in our register it speaks to

• Lethal autonomous weapons systems (LAWS)D11
• AI-enabled targetingD11
• Debate: ban, moratorium, meaningful human controlD11
• Right to human decision-makerD19
• Autonomous conflict escalationD23

Frameworks it relates to

• International humanitarian law (IHL)

How a reader might use it

• A school

  Use it to ask who still decides, when a system is faster than a person.

• A health system

  Use it as a model of insisting that a tool augment a clinician rather than replace judgment.

• A public body

  Use it as a model of a human-centred public position that names conflict, law, and care in one document.

Source

ICRC document page

International Committee of the Red Cross, Artificial intelligence and machine learning in armed conflict: a human-centred approach, 6 June 2019.