AI Can Clone Faces Instantly: Is There Cause for Concern?

Advanced Artificial Intelligence (AI) has simplified the process of creating digital doppelgangers.

For decades, AI tools have been programmed to generate faces. The face was chosen as a primary focus for testing capabilities due to multiple factors.

Firstly, the structure and layout of human faces are relatively consistent, making them ideal for early AI learning. There is also an abundance of facial images available online, providing more data for AI to train on.

Importantly, the choice to focus on facial recognition stemmed from its familiarity. Humans are highly skilled at recognizing and interpreting faces, while assessing the accuracy of AI-generated images of landscapes or buildings can pose challenges.

This is why studies showcasing AI’s ability to create hyper-realistic synthetic faces have continually garnered significant attention.

Consequently, AI has evolved into a facial recognition expert. Today, it can analyze live video footage, precisely identify facial features, and transition seamlessly between different pre-learned faces, including those of celebrities and public figures.

Moreover, users can adjust lighting and create exaggerated facial expressions. The “Deepfake” technology successfully mimics these effects.

Additionally, with a collection of images showing faces from various angles, AI has the capability to generate double the number of faces, enabling overlays on existing images. With such doppelgänger technology, anyone can effectively become you.

Best of all, this technology is freely accessible to anyone who knows how to install and operate it on a standard computer.

While creating a digital doppelgänger may require a certain skill level, it’s not difficult to find user-friendly software. However, these advancements raise significant concerns for privacy and verification.


This article addresses the inquiry (submitted by Julian Fisher via email) “How can I easily create my digital doppelganger?”

Please email us with your questions at Question @sciencefocus.com or reach out via Facebook, Twitter or Instagram Page (please include your name and location).

Explore our ultimate Fun fact and discover more fascinating science content.


read more:


Source: www.sciencefocus.com

Disappointing AI PAC-MAN clone receives negative reviews

tThere’s a lot happening in the world of video games and AI generators right now. Both Microsoft and Google have developed models that allow you to imagine a virtual world, albeit with some limitations. People are also using Grok, the Gen-AI chatbot from Elon Musk’s Xai, to create basic clones of old arcade games.

All you have to do is type “write.” On February 21st, Xai employee Taylor Silveira claimed to have created an exact replica of the 1980 Coinop Pac-Man using Grok 3. This version of Pac-Man includes dots, power pills, fruit, and all ghosts moving around the maze.

The key takeaway is that as long as the AI has the right commands in its software, it can generate a version within seconds. But how accurate are these Pac-Man clones really? Can just anyone create them? To answer these questions, I randomly approached some individuals in X who had posted their own Grok Pac-Man clones and asked them about their process.

I’ll give it a shot


John Hester’s Pacman. Illustration: Generated by AI

Author: John Hester, @hesterjohn
Time spent: 2 hours

First up, we have John Hester, a retired corporate software developer from California. When asked about Grok, Hester describes it as a large-scale language model (LLM) that competes with other models like ChatGpt and Llama. According to him, Grok 3 is the most advanced and continuously learns new things.

Hester’s experience with creating a Pac-Man clone using Grok was positive. He was able to give basic instructions to the AI and received a version of Pac-Man in seconds. Despite a few tweaks needed, Hester was impressed with Grok’s capabilities. He rates his experience 3 out of 5 stars.

Hester also mentions Elon Musk’s plans to use Grok for game development and emphasizes the cautious optimism surrounding AI advancements.

Score: 3 stars

Source: www.theguardian.com