If
we continue to allow AI systems to use human generated content...
https://scholarlycommons.law.wlu.edu/crsj/vol30/iss2/4/
Slavery.AI
The
artificial intelligence market is swarming. Supercharged start-ups,
global tech giants, and increasingly algorithmic governments target
diverse use cases with new and stunningly innovative AI applications
coming online every day. Where people are the computational subjects
of those algorithmic machinations, however, there is no law, present
or effective, to protect them against great and propagating harms.
Consequently, people become data production units, the commoditized
of the Data Industrial Complex and unfree, unpaid inputs to AI
production.
This
Article shares a new and provocative vision. It theorizes that
unregulated AI systems and uses are giving rise to an emergent form
of modern slavery: Slavery.AI. The Article examines the three
structural systems of power that were responsible for historical
chattel slavery and are at work today in Slavery.AI. Against these
interdigitating power structures, the evolution of two legal concepts
have brought forth, respectively, people-as-data-as-property and,
ultimately, as inputs to AI production, and modern slavery in all its
hideous permutations. At the confluence of these power systems and
trends, Slavery.AI is emerging, as defined, theorized, and
exemplified here. The Article crafts a crucible in which to test its
theory of Slavery.AI against the universal characteristics of systems
of slavery and demonstrates how those characteristics sounding in
property and in the abuse of power through cooptations of the rule of
law are firmly entrenched or on their way to being so. This
illustrated proof of concept holds. It also reveals that there may
be yet be opportunities for responsible leaders to save freedom and
to emancipate people from Slavery.AI.
From
sketch to AI “photo.”
https://www.researchgate.net/profile/Kalyani-Kute-3/publication/381584580_Forensic_Sketch_to_Real_Image/links/66755c981dec0c3c6f986f68/Forensic-Sketch-to-Real-Image.pdf
Forensic
Sketch to Real Image
In the realm of forensic
investigations, the reliance on hand-drawn sketches or verbal
descriptions to identify suspects or victims is well-established.
However, these traditional methods often present challenges due to
their inherent subjectivity and potential inaccuracies. Recognizing
the critical importance of enhancing the accuracy and efficacy of
forensic identification processes, our project delves into
cutting-edge techniques, particularly leveraging Deep Convolutional
Generative Adversarial Networks (DCGANs). Situated at the nexus of
image processing, artificial intelligence, and machine learning,
these advanced algorithms offer promising avenues for transforming
forensic sketches into remarkably realistic images.
Our project's primary objective is to
develop a robust Forensic Sketch to Real Image Conversion System
capable of generating highly authentic images from both hand-drawn
and computer-generated forensic sketches. By harnessing the power of
DCGANs, we aim to bridge the gap between the abstract representations
provided by traditional forensic sketches and the detailed, lifelike
images necessary for effective identification and investigation.
This innovative system holds immense potential to revolutionize
forensic practices, offering law enforcement agencies invaluable
tools to aid in criminal investigations, locate missing persons, and
address a myriad of forensic challenges.
Beyond its immediate applications in
law enforcement and forensic science, the Forensic Sketch to Real
Image Conversion System represents a significant stride towards
advancing the intersection of technology and justice. By providing
law enforcement agencies with the means to generate highly accurate
depictions of suspects or victims from rudimentary sketches, our
project seeks to bolster investigative capabilities while ensuring
fairness and accuracy in criminal proceedings. Moreover, the
potential societal impact extends beyond law enforcement, offering
hope and closure to families of missing persons and victims of crime
through improved identification methods