Our expertise

(Generative) AI

Generative AI is an advanced form of Artificial Intelligence capable of creating new data by learning from existing information much faster and more accurately than any human could. It feels like having a crystal ball that reveals the future based on present and historical data.

GenAI leverages the value embedded in your data. It has the capability to identify patterns and anomalies much faster and more accurately than any human could.

What we believe

But just because we can use generative AI, does not always mean we should. Understanding how AI models work, what and how they use the data, and the constraints they operate under is a prerequisite for using any AI solution. Especially generative AI.

Adoption is dependent on the level of trust people can have. Therefore, this must be baked into the design process of AI use cases.

Our six (Gen) AI design principles

To mitigate risks and bolster trust, we live by our (Gen)AI Manifesto.

This is a comprehensive set of design principles that ensure we keep ethics and trustworthiness top of mind, when developing (Gen)AI solutions.

Embedded legal compliance

We handle data in accordance with legislation such as GDPR and the EU AI Act. A legal review is an integral part of our (Gen)AI solution development to ascertain we clearly identify associated risks and take the necessary measures to ensure our solutions are Safe, Ethical, Transparent, and compliant with the EU AI act and other specific legislation such as the Machinery Directive.

Transparency and traceability

We create clarity about how models are trained and used. We train people in GenAI risks and safeguards.
Our developers are the epitome of awareness and help others grow their understanding of (Gen)AI and the boundaries and risks embedded in their data.
We help users understand how a model generates decisions. These decisions must be easy to understand and auditable.

Fairness and equity

We embed fairness, privacy, and safety into our designs. Limiting bias in AI outputs is a priority for all use cases. We ensure fairness and equity in outputs, by eliminating bias in the data that we use.

Responsible

Our solutions are developed in a socially responsible and sustainable way.
We prevent risks of physical or digital harm.
We consider the energy use of AI. We use our CO2 calculator to assess use cases and choose more sustainable configurations.

Accountable

We define who is responsible for AI outputs of use cases. Policies are in place to determine who is responsible for the decisions made with our AI solutions.

Resilient and secure

We build solutions that are resilient against outside influences and have security built-in as much as possible. We use our AI best practices, as we believe this prevents us from making unnecessary mistakes.

How we work

Our AI Center of Innovation has developed best practices to ascertain a focus on ethics and trustworthiness of the AI solutions we develop. They cover People, Process and Technology aspects of developing and managing (Gen)AI solutions.

Whether you don’t know where to start, lack the experience, or struggle finding the right use case, we can assist you in defining the feasibility of (Gen)AI use cases in 5-7 weeks.

Together, we explore, design, validate, and build a demonstrator that makes a use case tangible and supports a business decision to implement it.

When you are happy with the demonstrator and feel confident the value can indeed be realized, we are eager to assist you with its professional implementation.