Popular
Explore the future

Talk with your ancestors using AI

Time Machine 2002 Movie

A human sized avatar in the Time Machine movie was able to answer questions about anything, a very futuristic idea back in the year 2002. The character described himself as “a compendium of all human knowledge” which, fast forward 23 years later, sounds very much like ChatGPT. ChatGPT, however, can’t answer questions about topics that haven’t been publicly documented.

In this article I want to explore how artificial intelligence could answer questions about our ancestors. Or better yet, answer questions that our children and their children may have about us. Why does this matter and how could it be possible?

Check out the following scene from the 2002 Time Machine movie:

Human-like artificial intelligence character in the 2002 Time Machine movie.

why this matters

Growing up I enjoyed spending time with my grandparents and extended family to listen to their stories and the stories of their ancestors, which were also my ancestors. I learned lessons about courage, hard work and humility, as well as the negative effects of violence, addictions and other unfortunate events. These lessons were valuable insights as I portrayed myself into the future and started to navigate my own challenges.

I also learned how my ancestors came by boat from Europe and Africa to Honduras, and how some of them died in the attempt (cholera killed many of a dream at that time). I can only imagine what they went through leaving their home behind in the hope of a better future. I would pay to know what they were dreaming about and the different challenges they had to overcome.

Sharing our past with future generations can greatly inspire them with hope in the midst of life’s unforeseen events. It can also shed light on the importance of being ready to seize opportunities as they present themselves. Our life experiences as well as our personal values and beliefs can help give meaning to our descendants and their generations.

Some say that artificial intelligence will replace everything that is human in our society. But what is more human than tapping into the past to reveal lessons for the future? This is the basis on which all societies rely.

how it can be done

Now let’s discuss the steps needed to make this vision a reality. The technology stack needs 4 main layers as shown in the diagram below. Let’s go over each one of them.

1. The data layer

Data about us is constantly stored in servers around the world either purposefully or automatically. Automatically stored data can range from credit card payments to government information like passport travel stamps. Although this type of data can be interesting, for example if you need to prove that a stolen piece of art was rightfully purchased, it’s not much relevant for our future generations.

The most relevant data for future generations stems out of our own experiences, the relationships we have and the thoughts that go through our mind. I’m going to call this type of data, biographical data. A team at Microsoft patented a method for creating a conversational chatbot of a person based on their social media data, which is where we store most of this type of data today. However there can be more locations where we store biographical data, like for example our mobile phone storage, journaling notebooks, emails, etc.

Where we choose to store our biographical data matters since it contains sensitive information. Companies and open source software developers are spearheading the development of trust layers between data storage and AI with capabilities to anonymize contact and financial details, ensuring strict access controls and more.

Salesforce’s EinsteinGPT announcement of it’s AI trust layer.

I would like to see a company or governmental solution in the future that is dedicated to safeguarding human personal biographical data for people to transmit it as an inheritance to their children and others through a will or testament. If you know of a company with such a service please share it in the comments below!

2. The voice layer

My ancestors spoke different languages than the ones I speak today. One of them came from the city of Antwerp in Belgium where locals speak Flemish. Another one came from West Africa and would probably speak Akan or Fang. My native Honduran ancestors spoke Mayan and other Native American languages. I can’t help but wonder, where will my descendants live? Perhaps they’ll live in China, India or Mars?

It doesn’t matter how many languages will be spoken in the future, translation for AI is not a problem. It doesn’t matter where our descendants will migrate to and which languages they’ll speak, AI will speak to them and share our stories in their native language.

Apple’s iOS17 version introduced a feature to create your personal voice. This a step forward in being able to communicate with our descendants using our actual sounds and accent. This feature was especially built for people who are losing their voice for medical reasons and who would still like to communicate with others in their original voice. Users simply need to go through a series of questions and after some hours of processing, you can now enter a text and iPhone will read the text using your personal voice.

3. The video animation layer

The current state of the art in AI video animation includes different human avatars that are static in position and can read any text that you tell them to read, with a neutral personality (I will explain this later). Products like Synthesia or HeyGen make it easy to create video tutorials based on a script and a wide array of voices with different languages and accents to choose from. Check out the video below to get a glimpse of this fascinating technology:

Synthesia product demo showcasing a human-like avatar reading a script.

In the future I would like to see these types of videos render someone’s personality traits. What we need for that is a set of parameters to define how the avatar moves and makes facial expressions. These personality traits can be defined by artificial intelligence based on a person’s biographical data, using for example other videos of you, listening to your voice or simply reading stories written about you.

4. The AI layer

The artificial intelligence layer will have access to all the other layers in order to generate your interactive avatar. This avatar will be able to answer questions that your descendants may have and share stories about your life experiences and knowledge. In doing so, the avatar will interpret your personality based on your biographical data and generate the animation with truthful body movements, emotions, facial expressions and tone.

Imagine how beautiful it would be for these ancestor animations to answer questions about people’s life, experiences, and major events that happened in their generation.

The following animations are from the Deep Nostalgia project which uses AI to animate vintage photographs:

Do you have questions or comments? Leave a reply below!

Share this article
Shareable URL
Next Post

One year of generative AI

Comments 6
  1. What an excellent read, David!

    Really interesting and thoughtful take on the long term possibilities of AI.

    1. Thanks Stephen! The tendency is mainly to envision how AI can improve efficiency and productivity. Nevertheless it’s good to also envision how AI can enhance our human experience on this earth.

Leave a Reply

Your email address will not be published. Required fields are marked *

Read next

One year of generative AI

The article delves into the transformative impact of Generative AI in various aspects of life and work. It…
Happy new year 2024
Our site uses cookies. Learn more about our use of cookies: cookie policy

Subscribe to our Newsletter