Good Morning

Oct 02022
⪮ Good Morning :: ⪮ s01e10: People Potential
A photo using the new prototype Action Sampler Camera taking 4 photos each with a half-a-second delay.

AI Courtroom Sketches

The world of AI art is very young, but is already making a huge impact. Figuring out the ethics and use-cases are still not fully understood.

One excellent example we've seen is by Rachel Metz. She used Stable Diffusion to create a courtroom style sketch.

In instances where you are not allowed to take/publish photos (or screenshots), AI image creators can take a text prompt of what you see and create a convincing courtroom sketch.

Is AI a threat to more skilled, specialized jobs like courtroom sketch artists? Is there room for niche experts, or will we empower average people with even more, diverse, skills with black-box AI that we don't understand how it works?

Centaur Concept

The Centaur Concept is an idea from Matt Jones about human-machine collaboration.
7:29 video about What is the ideal scenario for human-machine collaboration?

A centaur is half-man, half-horse that work together. In today's analogy the horse is replaced by machine-intelligences.

The goal is not to get a computer to think like a human, but to use the tools that the machines excel at: unique inputs and sensors, speed and calculations, and unique outputs. Then create a seam and join the human and machine so they can be greater than the sum of their parts.

Computer Vision Dazzle Camouflage

CV Dazzle explores how fashion can be used as camouflage from face-detection technology, the first step in automated face recognition.

The original fashion patterns from 02010 worked, but the current DCNN face recognition system can now detect these faces.

New fashion styles are being created which will hide you from modern facial recognition cameras, but for how long?

Chernoff Faces

Invented by applied mathematician, statistician, and physicist Herman Chernoff in 01973, Chernoff faces display multivariate data using different facial features.

The bigger the eyes, nose, ears, etc., or the slant of the eyebrows or mouth are all encoding data. When each data point is displayed as a different facial feature type, we more clearly see the variation in faces as the output.

A People Pattern

Reaching out to people can be difficult and uncomfortable. If you take a people-centered approach when contacting someone new, you can bring satisfaction to both sides.

Use this as a starting point to improve your chances of a reply.