False Accusations of Cheating

“The students most susceptible to inaccurate accusations are likely those who write in a more generic manner, either because they’re neurodivergent, speak English as a second language or simply learned to use more straightforward vocabulary and a mechanical style. The result is that classrooms remain plagued by anxiety and paranoia over the possibility of false accusations.”

Read more at Bloomberg

31 AI & Data Science Articles from Oct 2024

Agentic AI is the top strategic technology trend for 2025 

MIT spin-off Liquid AI introduces more adaptive, less energy-hungry neural network models—inspired by microscopic worms

How The New York Times incorporates editorial judgment in algorithms to curate its home page

Deep Learning vs Data Science: What is more important, your data or your model?

AI Definitions: Large Language Models

Using Archetypes to Decode the Four Types of AI

4 ways to improve the retrieval of your RAG pipeline

Ways to exploit what AI can do to enrich data science solutions

Notion Templates Every Data Scientist Should Have in 2024

AI Definitions: Constitutional AI

AI-native software engineering may be closer than developers think

Replacing my Right Hand with A (a pitch for the “AI engineer“)

Geospatial intelligence capability communication is being weaponized (sub. req.)

AI Definitions: foundation models 

A Data Scientist GenAI Survival Guide

Maven helps Military emergency responders pinpoint where to place aid  

The 10 most prevalent and impactful vulnerabilities in large language model

AI Definitions: Data Scientist

An alternative framework for understanding memory in large language models

The release of the Qwen2.5-Coder series

Injecting Logic into Contexts for Full Reasoning in Large Language Models

CLIP for vision-language foundation models

AI Definitions: Washing

How I Would Learn Data Science in 2024

The Nobel Prize in physics goes to AI pioneers for their foundational work in neural networks & machine learning

Smaller LLMs perform well but this paper suggests they are also fragile

AI Definitions: AI model collapse

Correlating measures of hierarchical structures in artificial neural networks with their performance

How AutoML is changing the landscape of ML development

Can damage from fine-tuning an AI model be fixed? These researchers think so

The Data Centric Approach— rather than focusing on better models working on higher quality data

Insight into who will respond better in a crisis

A person’s capacity for healthy outcomes during difficulties is tied to their ability to define their life’s goals and values apart from the surrounding pressure to conform to a particular viewpoint.

In his book Generation to Generation, Edwin Friedman offers a way to test resistance to togetherness pressures, that is, possessing the power to say “I” when others are demanding “you” and “we.”

When presented with an issue that does not include “should” and “musts” some listeners will respond in a way that better defines themselves (such as “I agree” or “I disagree”). This person is likely to function well (emotionally) during a crisis. Other people may respond by attempting to define the speaker (comments like “How can you say that when…” or “After saying that I wonder if you are really one of us”). This indicates the person will likely resist progress toward healthy outcomes during crises and difficulties. People who more clearly define themselves are also more likely to take personal responsibility, whereas those who focus on the speaker are more likely to blame outside forces for their situations.  

One of the founding fathers of family therapy, Murray Bowen, suggested the capacity to define one’s own life’s goals and values apart from surrounding pressure, that is, to be a “relatively nonanxious presence in the midst of anxious systems” is an indication of taking “maximum responsibility for one’s own destiny and emotional being.” It shows up in “the breadth of one’s repertoire of responses when confronted with crisis.” The concept shouldn’t be confused with narcissism. For Bowen, differentiation means the capacity to be an “I” while remaining connected.

Stephen Goforth

18 Recent Articles about Students Using AI

Meet Sassy, the AI Chatbot Helping Students Find Their Dream Jobs – Ed Week 

How Students Can Use AI to Manage Their Time - CNET

Parents sue after student disciplined for using AI on school project in Massachusetts - CBS Boston

AI Detectors Falsely Accuse Students of Cheating—With Big Consequences – Bloomberg 

I write about AI for a living — and NotebookLM is the most exciting tech to arrive since ChatGPT – Tom’s Guide

The Students Who Are Overlooked by Most AI Tools – Ed Week  

Students with concentration issues turn to ChatGPT and similar AI tools, study finds -PsyPost 

Black teenagers twice as likely to be falsely accused of using AI tools in homework – Semafor  

A teacher caught students using ChatGPT on their first assignment to introduce themselves. Her post about it started a debate. – Business Insider

Kids who use ChatGPT as a study assistant do worse on tests - PopSci 

AI Cheating Is Getting Worse – The Atlantic

I tested 7 AI content detectors - they're getting dramatically better at identifying plagiarism – ZDnet  

Students and Professors Believe AI Will Aid Cheating – Inside Higher Ed 

Study shows disengaged students more likely to use AI tools for assignments – Phys.org 

Turkish student arrested for using AI to cheat in university exam – Reuters

AI can beat university students, study suggests - BBC

More than 400 Scottish students caught cheating using AI - AGCC 

What motivates students to use Generative AI and what would motivate them not to? – Dynamics of Writing

How to Pick the Best Leaders

“Employees who do well at their assigned tasks and score well on a simple IQ test are more likely to succeed as managers than noisy self-promoters. But there is an even better way to pick managers, according to these experts — directly test people’s aptitude for core management skills. The best managers, it turns out, are those who are actually good at one of the primary responsibilities of a manager — assigning the right projects to the right people.” - Inc

20 Recent Articles about How to Use AI

Surprising ways to prompt AI – Wonder Tools 

5 prompts to have a fun AI chatbot conversation - Mashable 

I write about AI for a living — here's how to become a true power user – Tom’s Guide 

Google unveils invisible ‘watermark’ for AI-generated text – Nature

Adobe promises AI tools that build 3D scenes, animate text, and make distractions disappear. – The Verge

Should You Be Nice to Your Chatbot? – Wall Street Journal

Adobe’s AI video model is here, and it’s already inside Premiere Pro - The Verge

I write about AI for a living — and NotebookLM is the most exciting tech to arrive since ChatGPT – Tom’s Guide  

Perplexity AI : How to Use It for Fast, Accurate Results – Geeky-Gadgets

Meta Unveils Instant A.I. Video Generator That Adds Sounds – New York Times 

I Built a Chatbot to Replace Me. It Went a Little Wild. - Wall Street Journal

Learn From My Worst AI Images and Fix These Biggest AI Fails – CNET 

AI's parent-teen knowledge gap – Axios  

Create Better AI Images With These Expert Prompt Writing Tips - CNET

How to use Midjourney's new AI image editor - Tom’s Guide 

How to cite ChatGPT in APA Style –  American Psychological Association 

How do I cite generative AI in MLA style? - Modern Language Association

What Is AI Best at Now? Improving Products You Already Own - Wall Street Journal 

Can Security Experts Leverage Generative AI Without Prompt Engineering Skills? – Tech Republic

Using AI to buy your home? These companies think it's time you should

You are who you are becoming

You are who you are becoming. Your virtue as a human individual is not related to any static, unchanging identity; it is about the person you are turning into—who you are today, as opposed to who you were yesterday, or could be tomorrow. You truly are, in Aristotelian terms, the life story you are writing through your actions and habits; as the historian and philosopher Will Durant summarized Aristotle’s view, “We are what we repeatedly do.” 

Research has consistently shown that when people see themselves as engaged in change and capable of progress, they are happier. You will have a better chance of realizing happiness if you can see yourself as a dynamic agent of your own progress. 

Arthur C. Brooks writing in The Atlantic

Matchless

To run yourself down hinders you doing what you can. In effect, when you belittle yourself, you are belittling God. He made you who you are – with your unique talents and lacks. To compare yourself with others is not good. Remember, you are God’s unique original! What an honor and privilege it is to be designed by the Almighty God! He know the end from the beginning. He never makes a mistake. He created you an individual – none other like you – for a purpose.

Ella May Miller

Hot New Job: AI Librarian

The AI revolution is creating demand for hot new job: AI librarian. “The growing demand for AI librarians “highlights how the evolution of technology is creating roles that merge traditional skills like information management with modern demands in data-driven environments. Companies need experts who can curate and translate the data into actionable insights.” -Digiday

16 Recent Articles about AI & Robotics

Future of farming? Carbon Robotics raises $70M for AI robots that blast weeds with lasers – Geek Wire  

The Battle Over Robots at U.S. Ports Is On – Wall Street Journal 

Microsoft is using AI-powered robots to help dismantle and destroy hard drives used in its data centers – Tech Radar

This AI humanoid robot helped assemble BMWs at US factory – Ars Technica  

A.I. Begins Ushering In an Age of Killer Robots – New York Times

We Need to Control AI Agents Now Automated bots are about to be everywhere, with potentially devastating consequences. – The Atlantic  

Is robotics about to have its own ChatGPT moment? – MIT Tech Review

An open-source vision-language-action model for robotics called OpenVLA has been released. – Venture Beat 

Ray Kurzweil is (still, somehow) excited about humans merging with machines – The Washington Post  

One-third of U.S. military could be robotic says former chairman of the Joint Chiefs of Staff – Axios  

Forget drones, this street-smart robot could be future of local deliveries – Fox News 

Sotheby's to auction its first artwork made by a humanoid robot – CBS News 

MIT engineers enabled robots to self-correct after missteps and carry on with their chores. – MIT Tech Review 

In America’s Factories, Even the Robots Are Getting Less Work – Wall Street Journal

The US Army is testing killer robot dogs with AI-powered rifles in the Middle East – Futuris 

AI and robots take center stage at ‘world’s largest tech event’ - CNN

AI Definitions: Large Language Models

Large Language Models (LLMs) - AI trained on billions of language uses, images and other data. It can predict the next word or pixel in a pattern based on the user’s request. ChatGPT and Google Bard are LLMs.

The kinds of text LLMs can parse out include grammar and language structure, word meaning and context (ex: The word green may mean a color when it is closely related to a word like “paint,” “art,” or “grass”), proper names (Microsoft, Bill Clinton, Shakira, Cincinnati), and emotions (indications of frustration, infatuation, positive or negative feelings, or types of humor).

More AI definitions here

Project Clean Machine

Adobe’s “Project Clean Machine” is an editing tool that “automatically removes annoying distractions in images and videos, like camera flashes and people walking into frames. For example, if a background firework causes a few seconds of the shot to be overexposed, Clean Machine will ensure the color and lighting are still consistent throughout the video when the flash itself is removed.”

More at The Verge

You are what you learn

If all you know is how to be a gang member, that's what you'll be, at least until you learn something else. If you become a marine, you'll learn to control fear. If you go to law school, you'll see the world as a competition. If you study engineering, you'll start to see the world as a complicated machine that needs tweaking.

I'm fascinated by the way a person changes at a fundamental level as he or she merges with a particular field of knowledge. People who study economics come out the other side thinking a different way from people who study nursing. And learning becomes a fairly permanent part of a person even as the cells in the body come and go and the circumstances of life change.

You can easily nitpick my definition of self by arguing that you are actually many things, including your DNA, your body, your mind, you environment and more. By that view, you're more of a soup than a single ingredient. I'll grant you the validity of that view. But I'll argue that the most powerful point of view is that you are what you learn.

It's easy to feel trapped in your own life. Circumstances can sometimes feel as if they form a jail around you. But there's almost nothing you can't learn your way out of. If you don't like who you are, you have the option of learning until you become someone else. Life is like a jail with an unlocked, heavy door. You're free the minute you realize the door will open if you simply lean into it.

Suppose you don't like your social life. You can learn how to be the sort of person that attracts better friends. Don't like your body? You can learn how to eat right and exercise until you have a new one. You can even learn how to dress better and speak in more interesting ways.

I credit my late mother for my view of learning. She raised me to believe I could become whatever I bothered to learn. No single idea has served me better.

Scott Adams, Dilbert.com