Identification AI systems and impactful algorithms

Determine whether your system is an AI system or impactful algorithm by answering a maximum of 8 questions.

* required

Automation of rules devised by humans

Does the system directly automate rules established in legislation, regulations, or other guidelines created by people?

If rules within your application are not explicilty predefined and documented, select here “no”. Software requirements with certain degrees of freedom are not considered to be rules devised by humans.

Which law or regulation?
*

Does the system contain a model or decision rules derived from data?

Which of the following categories does the system’s outcome fall into?

Provide a description of the output

Which of the following categories does the system’s outcome fall into?

Provide a description of the output
*

The following questions relate to the process in which the application is used. For the answers, it does not matter how automated this process is or what role the application has in the process. We will ask about the role of the application afterward.

Role in decision making

Are decisions made for individual citizens in the process? Note: a decision is much broader than a formal resolution.

Type of decision

What kind of decisions are made in the process? Choose the option that best matches the type of decision.

Describe the type of decision
*

Type of interaction with citizen

Does the process contribute to how the government categorises or approaches (groups of) residents?

Select “Yes” if you are uncertain, and briefly describe why.

Effect of the application

What is the effect of the system on the outcome of the process?

Choose the option that best fits.

Describe the effect of the system on the process
*

Based on your answers, your application is likely to be an impactful algorithm.

Your system is not an AI system or an impactful algorithm

Based on your answers, your application is likely to be an impactful algorithm.

A risk assessment for the AI Act must be filled in.

Newsletter

Stay up to date about our work by signing up for our newsletter

Newsletter

Stay up to date about our work by signing up for our newsletter

Building public knowledge for ethical algorithms