Honest cops?
https://brooklynworks.brooklaw.edu/blr/vol91/iss2/6/
Police and AI: When Abundantly Helpful Becomes Intrinsically Harmful
Artificial intelligence (AI) has rapidly crept into nearly all aspects of life, including in government, the criminal justice system, and policing. While Supreme Court Due Process jurisprudence has outlined certain boundaries for police interrogations, much police conduct is left for the states to regulate. Such regulation is sporadic and less restrictive than the public might assume, especially in the realm of police deception. Across jurisdictions, courts allow police to deceptively inform suspects that a witness identified the suspect of the perpetrator of a crime. That the suspect’s finger prints, DNA, or shoe prints were found at the scene of the crime. Police can even present fake evidence to suspects in an interrogation, including falsified lab reports, photograpghs, and more. With AI’s use expanding into law enforcement, there is a clear need to regulate police deception in interrogations before constitutional rights are infringed. This Note argues that while courts have long permitted various deceptive police tactics, the increasing sophistication and accessibility of AI tools pose unprecedented risks such as false confessions, bias, and potentially unwarranted public repremand. Through an analysis of case law, the evolution of Miranda and Due Process jurisprudence, and emerging AI applications in policing, this Note demonstrates how AI-enabled deception could exacerbate Due Process violations, undermine public trust, and increase wrongful convictions. It concludes by urging state legislatures to preemptively prohibit the use of AI to create false evidence in interrogations, advocating for a state-by-state legislative approach as the most effective means to safeguard constitutional protections in a rapidly evolving world.
Law is a Matrix?
https://scholarworks.uark.edu/arlnlaw/31/
Prompt Engineering For Lawyers: Blue Pill Or Red Pill: Hallucinations Risks And An Introduction To Prompt Engineering
In The Matrix, Neo’s choice between the blue pill and the red pill is essentially a choice between a comfortable illusion and an unsettling reality. Lawyers now face a similar decision with artificial intelligence. They can take the blue pill: ignore artificial intelligence or treat it like just another search engine, continuing a comfortable illusion that the new technology may not transform the practice of law. Or lawyers can take the red pill: acknowledge that artificial intelligence will transform the practice of law and learn how to use it competently, ethically, and effectively.
This Article is for those who choose the red pill. It begins with the problem of hallucinations, which makes blind reliance on artificial intelligence a professional hazard, and then turns to the first step in using artificial intelligence productively: understanding how it differs from Googling. When artificial intelligence is approached as a role-playing collaborator, such as a litigator, contract drafter, or judge, lawyers can enhance the accuracy, tone, and usefulness of the responses it provides.
Outside the box?
https://ojs.scipub.de/index.php/MSC/article/view/8331
THE PROBLEM OF THE CONSTITUTIONAL AND LEGAL REGULATION OF ARTIFICIAL INTELLIGENCE
This article examines the constitutional and legal problems arising against the background of the rapid development of artificial intelligence (AI), as well as the new realities generated by digital transformation. It offers a comparative analysis of the advanced constitutional practices of countries such as Chile, Greece, Mexico, and Brazil in the regulation of AI.
Referring to the theoretical concepts of prominent international scholars such as Lawrence Lessig, Frank Pasquale, and Mireille Hildebrandt, the article explores the principles of “code as law” and “legal protection by design.”
At the same time, it interprets the fundamental threats posed by AI in the spheres of algorithmic discrimination, the privacy of personal data, and neuro-rights.
The article proposes the application of a strict liability model within the civil law system of Azerbaijan for the compensation of damage caused by AI and suggests recognizing AI as an “autonomous source of risk.” In conclusion, it advances strategic solutions aimed at ensuring that national legislation evolves on the basis of the principles of digital constitutionalism and that the supremacy of human will over program code is preserved.
Maury Nichols points me to another interesting article.
Marcus asks AI chatbots various questions.
They seem entirely harmless. But they can tell the chatbots a lot about him.
Modern war.
The New Revolution in Military Affairs
How Ukraine is driving doctrinal change in modern warfare.
No comments:
Post a Comment