An algorithm similar to ChatGPT, known as “life2vec,” has been created by researchers from Nature Computational Science, demonstrating a 78 percent accuracy rate in predicting life outcomes such as lifespan and financial earnings.
Unlike other models, life2vec functions as a conversational chatbot, utilizing a person’s life story to anticipate future events.
Developed by scientists from Denmark and the United States, this algorithm serves as a repository for the knowledge of over 6 million Danish individuals. It incorporates diverse information, including details about income, occupation, residence, as well as incidents like injuries and pregnancies.
The uniqueness of life2vec lies in its ability to make predictions based on available details, resembling the conversational nature of ChatGPT. As a result, it stands on par with other algorithms designed for similar predictions of crucial life data.
MEN DIE
Previously, their efforts resulted in a model capable of generating estimates regarding the likelihood of premature death or the lifetime income of an individual. Various factors contributing to premature death encompass gender (specifically being male), a history of mental illness, or engagement in physically demanding labor.
Conversely, factors associated with extended lifespans include higher income levels and holding leadership positions. Life2vec operates by predicting the trajectory of a person’s life, treating the different stages akin to words in a sentence, much like the way ChatGPT generates text based on preceding input.
Similar to how ChatGPT responds to user prompts for song lyrics, poems, or articles, scientists can pose straightforward queries to life2vec, such as “Will this person pass away in four years?” regarding an individual’s life.
The model was trained using data spanning from 2008 to 2016. Leveraging population data, the system successfully forecasted over three-quarters of recorded deaths by the year 2020.
PEOPLE DON’T WANT TO KNOW WHEN THEY’RE GOING TO DIE
In adherence to privacy principles, the participants’ data has not been disclosed thus far, and scientists are diligently working to disseminate the results without compromising confidentiality.
The entirety of the data utilized in the project originates from Denmark, implying that the estimates may not be universally applicable to individuals residing in other locations. Additionally, the project officials note that the majority of people may not have an interest in knowing the precise timing of their demise.
You may also like this content
- What is Artificial Intelligence?
- Discover the Top 15 Artificial Intelligence Tools and Websites
- Understanding How Artificial Intelligence Works
-
-
-
-
-
-
-
-
-
-
