07/03/2025

Visgraf video combines avatars of Aristotle and Steve Jobs.

Avatares de Steve Jobs e Aristóteles

“The vision is to create systems that empower individuals while nurturing wisdom and ethical understanding.” This phrase was part of the conversation between the avatars of the Greek mathematician Aristotle (384 BC – 322 BC) and the American businessman Steve Jobs (1955 – 2011). The conversation was made possible thanks to the application of multicasting technology, developed by IMPA researcher and leader of Visgraf (IMPA's Computer Graphics Laboratory), Luiz Velho, in partnership with researcher Matteo Moriconi (VFXRio). The video, as well as other research results, were presented during the lecture “From Aristotle to Jobs for Celestia Borealis” on Wednesday (26) at IMPA.

“What’s really interesting are these kinds of experiences, where we have multiple participants, characters driven by artificial intelligence. We can have actors, directors, any kind of experience. It can be a teaching experience, it can be a dissemination experience, it can be an artistic experience,” Velho said.

The presentation was part of a series of six lectures, available on Visgraf's YouTube channel , about the research, which aims to understand the interface between humans and machines, as well as the evolution of the digital environment.

The main inspiration for the researchers' most recent work was Steve Jobs' proposal to capture, through computer systems, a worldview, a persona, or an idea. Inspired by this, Velho and Moriconi explored the potential of generative AI with new media and narrative systems, as well as working on data capture by AI.

In addition to the conversation between Aristotle and Jobs, the researchers also presented new tests conducted with the Celestia Borealis robot , an artificial intelligence trained to incorporate personality in handling requests. “We launched Celestia some time ago and we are evolving its photorealism. From the first experiments until now, the idea is to make it as photorealistic as possible, but also to be able to transport it to immersive and interactive experiences not only with humans, but also with other avatars or with the environment itself,” explained Moriconi.

The tests and research advancements are being conducted using Multicasting/Multimodal technology. “The Multicast Multimodal System is a playground, an experimental space for generative AI, for building systems of this type. We are working with a creative platform for the ideation, development, construction, and realization of media experiences using AI. The basic idea of the AI tool we use is, first, a system that is distributed, uses internet communication, uses all kinds of new media, including those that haven't yet appeared, to create an expanded reality. So, this includes our real reality, virtual reality, augmented reality, mixed reality,” Velho explained.

Looking to the future, researchers want to continue exploring new possibilities through augmented reality and AI capabilities. The idea is to provide more flexible interactive experiences, which could be, for example, in facilities or on social media.