Perspective.

https://www.foreignaffairs.com/united-states/ai-weapons-and-dangerous-illusion-human-control

AI Weapons and the Dangerous Illusion of Human Control

… Although it’s comforting to imagine that living, breathing people will restrain mindless algorithms from killing indiscriminately, this consensus is out of step with technological reality. The AI models driving contemporary autonomous weapons are often too sophisticated for even the best-trained operators to supervise. What is more, every proposed use for autonomous weapons imagines them operating at massive scale, which would require levels of data, speed, and complexity that would make meaningful human intervention impossible. Under normal conditions, expecting a human being to weigh the merits of an AI system’s analysis and suggest courses of action would be challenging. Under combat conditions—marked by severe stress, limited time, and interrupted or nonexistent communications between individuals, units, and higher authorities—doing so would be impossible. Rather than indulge the illusion that humans will have the capacity to control autonomous weapons in wartime, therefore, militaries must build trust in their models for autonomous weapons now—under peacetime conditions—and let them function without excessive human intervention when the shooting starts.

This is a first. Is it possible they have proof but are unwilling to admit their security failure?

https://www.cnn.com/2024/12/06/europe/romania-annuls-presidential-election-intl/index.html

Romania’s top court annuls presidential election result

Romania’s constitutional court has annulled the first round of the country’s 2024 presidential election process, which was narrowly won by far-right ultranationalist candidate Calin Georgescu amid allegations of Russian interference.

The historic decision means a second-round runoff that was scheduled for Sunday, when Georgescu would have gone head to head with his centrist rival Elena Lasconi, will no longer take place.

In a press release, the court said the annulment was based on Article 146(f) of the Constitution, emphasizing the need to ensure the correctness and legality of the electoral process.

… It also comes just days after the vote was hit with fears over foreign interference, after declassified documents from Romania’s top security council found evidence of “aggressive hybrid Russian attacks.”

Not quite Skynet, but getting closer.

https://www.cnbc.com/2024/12/04/openai-partners-with-defense-company-anduril.html

OpenAI partners with defense company Anduril

OpenAI and Anduril on Wednesday announced a partnership allowing the defense tech company to deploy advanced artificial intelligence systems for “national security missions.”

… The OpenAI-Anduril partnership announced Wednesday will “focus on improving the nation’s counter-unmanned aircraft systems (CUAS) and their ability to detect, assess and respond to potentially lethal aerial threats in real-time,” according to a release, which added that “Anduril and OpenAI will explore how leading edge AI models can be leveraged to rapidly synthesize time-sensitive data, reduce the burden on human operators, and improve situational awareness.”

Anduril, co-founded by Palmer Luckey, did not answer a question about whether reducing the onus on human operators will translate to fewer humans in the loop on high-stakes warfare decisions.

Perspective.

https://insight.kellogg.northwestern.edu/article/century-of-ai-innovations

Overnight Success? AI Has Been a Century in the Making.

AI’s emergence in recent years may appear sudden to some, but its development has been years in the making. Professor of finance Sergio Rebelo digs into AI’s past through the lens of a macroeconomist to uncover four lessons that can help people prepare for AI’s growing role in society: recognize that success can take decades; let intuition guide people toward smart risks; stay the course through challenges; and mind the hype.

Useful for non-lawyers too?

https://www.bespacific.com/the-right-way-to-use-genai-to-draft-a-brief/

The Right Way to Use GenAI to Draft a Brief

The Right Way to Use GenAI to Draft a Brief, Carolyn Elefant (December 2024) – “It’s been two years since the public release of ChatGPT and lawyers are still using it improperly and facing sanctions, as reported last week in Reuters about an incident in the Eastern District of Texas Instead of wringing our hands or using stories like these to hawk pricey commercial products (spoiler alert: the lawyer claimed he used a feature on LEXIS AI to verify citations), let’s learn how to use genAI correctly. Methodology: The reported E.D. Texas hallucination case involved this response in opposition to this Motion for Summary Judgement which I used as the basis for this exercise. I don’t like to criticize other lawyers’ work because we never know what was going on behind the scenes (non-paying client or personal struggles), but let’s just say that even without the handful of AI inaccuracies, this response could use some improvement. Here’s how I would have used Claude (the lawyer’s preferred AI tool) to respond to the SJ Motion…”

Human error or out of control AI?

https://www.huntonak.com/privacy-and-information-security-law/recent-texas-case-highlights-increasing-relevance-of-privacy-and-security-laws-to-e-discovery-process

Recent Texas Case Highlights Increasing Relevance of Privacy and Security Laws to E-Discovery Process

On November 6, 2024, a Texas state district court jury found that a large e-discovery vendor violated Title 7, Chapter 33 of the Texas Penal Code, which provides that accessing a computer without its owner’s permission is a Class B misdemeanor. This case highlights the importance for e-discovery vendors of considering data privacy and security requirements in the course of discovery proceedings.

The court awarded $50,000 in damages against the vendor, finding that the vendor ignored e-discovery protocol when accessing an individual’s personal email account to search for emails containing 12 search terms dating back to 2012. As a result, the vendor downloaded more than 34,000 emails from the individual’s email account that included sensitive information such as medical data, Social Security numbers and attorney-client privileged information. According to reports, if the vendor had not deviated from the e-discovery protocol, only 600 relevant emails would have been in scope of discovery.

Data minimization is a key feature of many privacy laws, including the recently enacted Texas Data Privacy and Security Act, and companies are generally required to limit the collection of personal data to what is adequate, relevant and reasonably necessary to the disclosed purpose for which the personal information will be processed.

Law for geeks?

https://www.zdnet.com/article/why-this-long-awaited-os-merger-muddles-dojs-case-against-google/

Why this long-awaited OS merger muddles DOJ's case against Google

Google's merger of ChromeOS and Android could transform ecosystems and complicate antitrust remedies while challenging regulators to balance innovation, competition, and consumer impact.

Can anyone be ‘smart’ under this formula?

https://www.psychologytoday.com/intl/blog/the-digital-self/202411/what-does-it-mean-to-be-intelligent-today

What Does It Mean to Be Intelligent Today?

In 2017, I explored the concept of Technology Quotient (TQ) as a measure of our ability to navigate the evolving digital world. The conversation was centered on a dichotomy: technology’s battle between IQ and EQ. Back then—a technological lifetime ago—TQ was framed as a way to capture our capacity to understand, interact with, and thrive in an increasingly technical society—a society still dominated by hardware, software, and user engagement.

Fast forward to today, and the world has shifted significantly. Large Language Models have reshaped the technological and human landscape, pushing us beyond my former perspective on TQ. These tools blur the lines between intellect, emotion, and creativity. They are not just technical systems we operate; they are collaborators that amplify human potential and even thought.

In light of these changes, it’s time to revisit and redefine my earlier thinking about TQ—not as a battle between IQ and EQ, but as a multidimensional framework that captures the symbiosis between humans and AI. This new TQ reflects the skills and literacies required to thrive in a world where technology is not just a tool, but an extension of our cognitive and even our emotional capabilities.

Could Hollywood create a ‘star’ from whole cloth?

https://variety.com/vip-special-reports/generative-ai-celebrity-deepfakes-digital-replicas-special-report-1236218083/

Generative AI, Celebrity Deepfakes & Digital Replicas: A Special Report

Generative AI represents a double-edged sword for talent in the entertainment industry, bringing creative and monetization opportunities but also challenging risks.

This special report from Variety Intelligence Platform offers a deep dive into how name, image, likeness and voice (NILV) of actors and artists will be protected from misuse, as talent agencies and employers aim to combat the online spread of nonconsensual celebrity deepfakes and synthetic content.

Meanwhile, AI digital replicas are also becoming available to exploit for celebrities’ creative and commercial use. As authorized AI versions of talent likeness increasingly appear in content, celebs will need assurances that technical systems are in place for them to engage safely and benefit from their digital replicas to create new content experiences.

Tools & Techniques. Because you never know when these might be handy.

https://www.bespacific.com/bookmark-calculator-soup/

Bookmark ‘Calculator Soup’

Lifehacker – For the Next Time You Need to Convert Just About Anything – “We all have a few go-to calculations that we find ourselves doing often, like converting ounces to cups or miles to kilometers—that sort of thing. When I need to do a quick conversion, I use Calculator Soup, a simple website offering hundreds of purpose-built calculators for things like inflation, unit conversion, gas mileage, algebra, and so much more. The site can be overwhelming at first glance, with its old-school aesthetic and list of categories, but it’s full with the kinds of tools you’d otherwise search for and find ad-ridden versions of. For example: if you’re trying to get the best possible car loan there’s a car loan calculator …”

Seems backward to me. Is this the slipperiest of slopes?

https://www.bespacific.com/feds-can-film-your-front-porch-for-68-days-without-a-warrant-says-court/

Feds Can Film Your Front Porch for 68 Days Without a Warrant, Says Court

Gizmodo: A federal court says your privacy is diminished due to the proliferation of video cameras throughout society. “…The federal court’s decision says that video cameras have become “ubiquitous,”  and have therefore diminished our expectations of privacy. Police officers wear body cameras now, cellphones have cameras, and many doorbells record your porch. The court isn’t wrong that cameras are everywhere. However, law enforcement has a long history of blurring the lines of privacy with modern recording technology.  Politico detailed how Ring handed over a full day’s worth of camera footage against a man’s will, in order to convict his neighbor of a crime. The network of Ring cameras also was used by law enforcement for years to obtain footage of criminals without search warrants …”

See also Hartzog, Woodrow and Selinger, Evan and Gunawan, Johanna, Privacy Nicks: How the Law Normalizes Surveillance (March 10, 2023). 101 Washington University Law Review 717 (2024), Available at SSRN: https://ssrn.com/abstract=4384541 or http://dx.doi.org/10.2139/ssrn.4384541

Is AI ever a better choice?

https://www.infodocket.com/2024/12/01/research-article-preprint-suspected-undeclared-use-of-artificial-intelligence-in-the-academic-literature-an-analysis-of-the-academ-ai-dataset/

Research Paper (Preprint): “Suspected Undeclared Use of Artificial Intelligence in the Academic Literature: An Analysis of the Academ-AI Dataset”

Since generative artificial intelligence (AI) tools such as OpenAI’s ChatGPT became widely available, researchers have used them in the writing process. The consensus of the academic publishing community is that such usage must be declared in the published article. Academ-AI documents examples of suspected undeclared AI usage in the academic literature, discernible primarily due to the appearance in research papers of idiosyncratic verbiage characteristic of large language model (LLM)-based chatbots. This analysis of the first 500 examples collected reveals that the problem is widespread, penetrating the journals and conference proceedings of highly respected publishers. Undeclared AI seems to appear in journals with higher citation metrics and higher article processing charges (APCs), precisely those outlets that should theoretically have the resources and expertise to avoid such oversights. An extremely small minority of cases are corrected post publication, and the corrections are often insufficient to rectify the problem. The 500 examples analyzed here likely represent a small fraction of the undeclared AI present in the academic literature, much of which may be undetectable. Publishers must enforce their policies against undeclared AI usage in cases that are detectable; this is the best defense currently available to the academic publishing community against the proliferation of undisclosed AI.

Direct to Full Text Article

No good way?

https://www.ft.com/content/e2fa34b2-6987-494d-a81a-1bdb6693671f

The legal battle against explicit AI deepfakes

It is easier than ever to forge graphic videos and images. But campaigners hope that new laws could offer a template for controlling artificial intelligence