The rise of digital avatars | From Ruud Gullit to your organization

Well-known footballer Ruud Gullit has joined a growing list of public figures who are creating digital versions of themselves. On RTL Tonight, he showed his AI clone: a virtual version of Ruud Gullit trained on the way he speaks, thinks and responds.

https://www.youtube.com/watch?v=ArwmkxX97qA

Gullit expects this to become completely normal within a few years. And honestly, I think he may well be right.

We are not the only ones seeing this trend. People such as LinkedIn co-founder Reid Hoffman and Deepak Chopra have also created digital versions of themselves. Reid Hoffman uses his digital twin to make his ideas and presence scalable, for example across multiple languages. Deepak Chopra has an AI platform that allows people to ask questions about wellbeing, mindfulness and personal growth through a digital version of him.

But why are people doing this? And why do they believe this is the future?

Your knowledge available 24/7, in every language

I believe every person has unique qualities. One person may be strong socially, another commercially, technically or creatively. With a digital avatar, you can make that knowledge, experience and personality available much more broadly: 24 hours a day, 7 days a week, to people all over the world.

Of course, you could say: “Isn’t that what a website is for?” But a website is often limited. Not everything is available in every language, the interaction is usually static, and the depth of the content remains limited.

Take, for example, a heating engineer with twenty years of experience. In practice, they have encountered thousands of situations: faults, installations, exceptions, customer questions and solutions that are not found in a standard manual. It is almost impossible to put all that knowledge properly on a website.

With a digital avatar, this works differently. It can be fed with your knowledge, experience, documents, videos, tone of voice and way of explaining things. Combined with a language model, such an avatar can answer not only general questions, but also much more specific ones. Provided, of course, that the avatar has been set up properly.

A digital avatar can also have social value

A digital avatar does not have to be purely business-oriented. A digital twin can also have social value. Think of people who feel lonely and would like to speak to a familiar face or recognizable voice.

The big difference compared with a regular chatbot is that someone does not only see text, but also hears and sees a face speaking. You can increasingly give a digital twin your own voice, character, way of speaking and communication style. And when conversations are stored safely and responsibly, the avatar can also build on previous interactions in a future conversation.

That makes the experience more personal than a standard chatbot. At the same time, it is important to remain honest: a digital avatar is not a human being, but we are getting increasingly close to a likeness.

How do you create a digital avatar of yourself?

With a bit of technical knowledge, almost anyone can now create a simple digital avatar of themselves. But to truly capture your voice, character, knowledge and way of thinking, more specialist expertise is often needed.

A commonly used combination consists of tools for visuals, voice, hosting and artificial intelligence. Think, for example, of HeyGen for the visual avatar, ElevenLabs for voice technology, an online server environment such as AWS, and a language model such as OpenAI or another LLM platform.

HeyGen for your visual avatar

HeyGen specializes in creating AI videos and visual avatars. You can upload a photo or video of yourself and use it to create a digital version that can speak and move.

A video usually produces a more natural result than a photo alone, because the system then has more information about your facial expressions, lip movements and mimics. The better the recording — good lighting, a steady image and clear speech — the more realistic the end result.

ElevenLabs for your voice

ElevenLabs is widely used for voice cloning, text-to-speech and voice agents. With a good, clean recording of your voice, a digital voice can be created that comes increasingly close to your own voice.

Through connections with other systems, that voice can then become part of a larger digital assistant. Think of holding conversations, answering questions or being connected to other software. Tasks such as email, WhatsApp or phone calls can easily be built in.

AWS to make everything work together

For many people, the technical side is the biggest hurdle. You need to connect different systems: the visual avatar, the voice, the language model, your own knowledge base and possibly your website or app.

An online server environment such as Amazon Web Services can help with this. With the help of tools such as ChatGPT or Claude, you can get much further today than a few years ago, even without a large technical team. Still, it remains important to set this up properly and securely, especially when working with personal data, customer questions or commercial applications.

OpenAI or another LLM as the brain

A digital avatar also needs a language model: the “brain” that understands what someone is asking and formulates an answer. OpenAI, the company behind ChatGPT, is one of the well-known providers of such language models, but there are also other providers.

Which model is most suitable depends on your goal. Some models are cheaper, while others are stronger in reasoning, speed, multilingual use or natural dialogue. Because this technology is developing rapidly, the best choice a few months from now may already be different.

From experiment to practical application

What stands out to me most is that digital avatars are no longer just a futuristic experiment. The individual building blocks already exist: visuals, voice, language models, knowledge bases and integrations with other systems. The real challenge is no longer only technical, but above all about how to apply this properly, safely and credibly.

Because a digital avatar should not only be able to talk. It has to fit the person or organization behind it. The tone needs to be right, the knowledge needs to be reliable, and users need to understand when they are interacting with AI. That is exactly where the difference lies between a fun demo and an application that truly adds value.

For entrepreneurs, experts and organizations, this may become increasingly interesting in the coming years. Not because everyone needs a digital clone of themselves tomorrow, but because knowledge, availability and personal communication are becoming more scalable.

At Mindd, we are following this development closely and working on ways to make these kinds of AI applications practically usable. Not as a gimmick, but as an extension of knowledge, service and communication.

Ruud Gullit’s digital avatar mainly shows where we are heading. The question is not whether this technology will improve, but how we use it in a way that remains human, transparent and valuable.