Skip to content
BENIBIX Learn

DPIA vs AI Impact Assessment

People think a DPIA and an AI impact assessment are the same check, and they are not. A DPIA looks only at the risks of using personal data. The GDPR asks for it. An AI impact assessment looks at every way an AI system could harm people, such as bias or safety. It counts even when no personal data is used. One AI project can need both.

Data🔒 Data Privacy

Data Protection Impact Assessment (DPIA)

A risk check you write first. Then a risky data use may start.

A data protection impact assessment is a written risk check before a new use of personal data. Article 35 of the GDPR asks for one when a use is likely to be high risk for people. Tracking people on a large scale, scoring them or using health data are examples. The check describes the use and asks if it is really needed. Then it lists the risks and the steps that cut them. If a high risk is left, the firm must ask the watchdog before it starts.

Example
A gym wants members to get in with a face scan. Its DPIA flags face data as sensitive. So the gym keeps a card as another way in, and deletes the scans of anyone who leaves.
Why it matters
It finds the harm on paper, while the plan can still change. Skipping a required one can bring a fine even if nothing goes wrong.
Full entry
AI⚖️ Ethics & Regulation

AI Impact Assessment

A written check of how an AI could hurt people. It is done before launch.

An AI impact assessment is a written check, done before launch, of how an AI system could affect people. It asks who will use it, who it makes choices about, and what could go wrong for them. It looks at bias, privacy, safety and people's rights. Then it lists the steps to cut each risk, and who owns each step. Under the EU AI Act, some users of high-risk systems, such as public bodies, must do one for people's basic rights.

Example
A city plans an AI that ranks which homes to inspect first. Its team writes an impact assessment and finds poor areas would be checked far more. They change the model before launch.
Why it matters
It is far cheaper to find a harm on paper than after it has hurt people.
Full entry

In one line

Data Protection Impact Assessment (DPIA)
A risk check you write first. Then a risky data use may start.
AI Impact Assessment
A written check of how an AI could hurt people. It is done before launch.

These live in different corners of the glossary (Data and AI), which is exactly why they get confused.