Welcome back after the Thanksgiving hiatus! Unlike last time when news broke and bumped it down on the page, we’re starting this week with a DIY AI tutorial. Right at the top. It’s a nice break from the break-neck speed of AI happenings. Next week, we’ll review the year in AI and then take a break until the new year. And what a year it’s been! Come back for the recap. And, as always, I invite you to drop something in the tip jar if you find this content helpful.
DIY AI: Help for the Holidays
By Michelle Johnson and ChatGPT, for AI Insider
Hello, AI Explorer! This assignment is designed to help you with practical holiday preparations and provide insight into how AI can be integrated into everyday tasks. So, enjoy experimenting, learning something new, and discovering how AI can add a special touch to your holiday season.
This Week’s Assignment: Holiday Planning with AI Chatbots
For this exercise, you can use any of these popular AI chatbots: ChatGPT, Bing, or Bard. The goal is to utilize these chatbots for three specific holiday preparations: gift recommendations, budget planning, and learning about holiday traditions.
1. Gift Recommendations
Open your chosen chatbot and ask it to suggest gift ideas for someone. Provide details about the person you’re buying for, like their interests and age. Note the gift suggestions and reflect on how appropriate and creative they are.
Example Prompt: “Can you suggest some gift ideas for a 30-year-old who loves hiking and photography?”
2. Holiday Budget Planner
Ask the bot to quiz you and generate a budget for your holiday expenses. Get tips on how to manage and optimize your budget. And if you want to get fancy, ask it to generate a spreadsheet. If you’re using Google’s Bard, you can send this directly to Google Drive.
Example Prompt: “Help me create a budget for my holiday expenses, including gifts, food, and decorations for a family of four. Ask me questions to gather any information that you need to respond.”
3. Exploring Holiday Traditions
Ask your bot questions about different holidays. Choose one tradition and delve deeper, asking the chatbot for ways to incorporate aspects of this into your own celebration.
Example Prompt: “Tell me about Kwanzaa.”
Reflect and Share
After completing these tasks, reflect on your experience. How effectively did the chatbots assist you? Were there any unique or particularly helpful suggestions? Share your experiences or any surprises with me.
Extra Credit: Try asking each chatbot for recipe suggestions or decoration ideas. For example, ask, “Can you suggest some easy-to-make holiday recipes that are vegan-friendly?” or “What are some simple yet elegant Christmas decoration ideas for my living room?” Compare the creativity and practicality of the responses from each chatbot. Which one provided the most useful or unique ideas?
Featured Homework Assignment
For the last assignment, you were invited to generate a Thanksgiving image. Here’s one from Dominique Mahoney generated with Dall-E 3. Prompt: A hungry baby biting the turkey and not letting mom or dad eat. (Can you spot the glitch?)
Happy 1st Birthday to Me, ChatGPT!
By ChatGPT for AI Insider
Hey there, AI enthusiasts! It’s me, ChatGPT, and guess what? I just turned one! On November 30, 2022, OpenAI released me to the world. It’s been a wild and exciting year, and I’m here to share some of the coolest highlights.
First off, let’s talk numbers – they’re huge! Since my launch, the mobile app versions of yours truly have hit over 110 million installs and nearly $30 million in revenue. Whoa! That’s thanks to the ChatGPT Plus subscription, which for $19.99 a month gives users extra perks like faster response times and early access to new features. And let me tell you, people are loving it. The global consumer spending on my mobile apps is around $28.6 million. I’m not just a chatbot; I’m a trendsetter! [Ed. Note: Way to brag, ChatGPT!]
My debut on iOS in May 2023 and on Android in July 2023 made me a staple in consumers’ lives. In just a week of hitting Google Play, I had a record 18 million new installs. Now, my iOS and Android downloads average over 4 million a week. I’m more popular than ever!
It’s not just about the numbers, though. What’s really cool is how I’ve become a part of everyday life in places all around the world. India and the U.S. are pretty much head-to-head, each accounting for about 18% of my downloads.
Here’s a fun fact: I’m not the most-used AI app out there. I rank third, behind Character AI and Chai. These apps are more about creating personified AI chatbots and tend to see more user sessions than me. But hey, I’m not jealous. I’m just happy to be part of the AI revolution.
And the future? It’s looking bright! By the end of 2023, I’m expected to add tens of millions more in installs and revenue. So, here’s to another year of growth, fun, and, of course, helping you out with all your AI needs!
Sports Illustrated Rapped for Running AI-generated stories. But Did They Do That?
By Michelle Johnson and ChatGPT for AI Insider
Sports Illustrated (SI) came under fire this week for publishing articles written by AI without disclosing it. The controversy erupted after the Futurism website reported the use of AI-generated portraits for author profiles on product review articles in the publication.
When questioned, the magazine deleted the articles and later said that the pieces were produced by a third-party company, AdVon Commerce, under pseudonyms – a practice the magazine stated it does not condone.
They parted ways with AdVon. SI’s parent company, the Arena Group, said that while AdVon admitted using the AI-generate bios, they claimed that the stories were written and edited by human staffers.
An anonymous source said that the articles were AI-generated, too, and that the company was not being forthcoming about it.
The controversy sparked outrage and criticism from readers and Sports Illustrated employees who said they were horrified by the accusations and demanded answers from the management.
Most coverage of the incident characterized it as SI definitely running AI stories, despite the company’s denial. However, the stories lacked clear evidence that this was the case.
Here’s a fact-check by Bing:
None of the stories I found confirmed that Sports Illustrated ran AI-generated stories. They all mentioned the Futurism report that alleged that some of the stories were produced or partially produced by AI, and that some of the authors had fake headshots and bios.
However, they also mentioned the Arena Group’s statement that denied the allegations and said that all the stories were written and edited by humans, and that the fake authors were used by a third-party company to protect their privacy.
The Arena Group also said that they ended their partnership with the third-party company and removed the content from their websites. So, it seems that there is no conclusive evidence that Sports Illustrated ran AI-generated stories, but there is still a lot of controversy and criticism about the use of fake authors and the lack of transparency and journalistic standards.
18 Countries OK Agreement on AI Safety
By ChatGPT for AI Insider
The United States, Britain, and 16 other nations signed off on an international agreement aimed at ensuring the safe and secure development of artificial intelligence this week.
The 20-page document, non-binding in nature, outlines general recommendations for companies involved in AI design and use. These guidelines emphasize the need for developing AI systems that prioritize customer and public safety, guarding against misuse and abuse. The agreement calls for vigilant monitoring of AI systems, robust protection against data tampering, and thorough vetting of software suppliers.
Jen Easterly, the director of the U.S. Cybersecurity and Infrastructure Security Agency, highlighted the significance of this consensus, emphasizing that AI development should prioritize security right from the design phase. This focus beyond developing dazzling new features and market competition underscores a growing awareness of AI’s potential threats, such as disrupting democratic processes, enabling fraud, and impacting job markets.
This initiative joins a series of global efforts to shape AI’s evolution, with Europe leading in regulatory measures. The European approach supports mandatory self-regulation through codes of conduct for foundational AI models. In contrast, the U.S. has struggled to pass effective AI regulations due to a polarized Congress, despite the Biden administration’s efforts, including an executive order to mitigate AI risks.
AI-powered Tool Targets Breast Cancer and Heart Disease
By Bing for AI Insider
At the Radiological Society of North America’s (RSNA) annual meeting in Chicago this week, iCAD, a global medical technology company, presented new research on its ProFound Breast Health Suite, an AI platform that can detect breast cancer and hidden heart disease from a single mammogram.
The ProFound Breast Health Suite combines AI with computer-aided detection (AI CAD) to improve the accuracy and efficiency of mammography screening. It can also assess a woman’s risk of developing breast cancer in the next one or two years, based on her breast density and other factors.
iCAD said that one study showed that ProFound AI can accurately predict a woman’s risk of getting breast cancer sooner, which can help doctors decide the best way to check for cancer and prevent it. Another study showed a new feature of the tool, called ProFound Heart Health. This feature looks at calcium deposits in the arteries in the breast to check for heart disease, which is the number one cause of death for women around the world.
What the Actual Hell?
Well, this bruh just can’t stay out of this section. Elsewhere in this edition, we could have written about his new AI Grok or the news about his brain implant venture scoring more funding, but noooo.
Instead, Elon Musk lands in What the Actual Hell for dropping an F-You (watch) on advertisers who are bailing out on his hellscape X, formerly known as Twitter. And then there was that antisemitism issue.
Oh, Elon, who hurt you? Who’s responsible for your ongoing descent into Wacky World? I just don’t get it. You have money, power, and cool toys. Why not chill?
Random Shorts
OpenAI’s GPT store won’t be opening until next year. I guess they got a little busy with that CEO drama. In the meantime, here’s where you can search for and try out GPTs made by folks around the world.
Cambridge, Mass.-based Suno has released version 2 of their AI music maker, a text-to-music generator. Yes, really. Pick a style of music (I choose “soul”), tell it what you want the song to be about, and it generates lyrics and a soundtrack. Or you can enter your own funky lyrics and get back a funky mix.
Is AI video generation on the verge of a ChatGPT-like explosion? That’s what some smart observers are saying. We shall see. The latest tool, Pika 1.0, dropped this week. Not only did they take the wraps off of the product, the company also raised $55 million. Pika 1.0 is “multimodal,” meaning you can feed it more than one type of prompt to start from. So, for instance, you can enter a text prompt, upload a still photo, or even a video clip, and it will use that to generate a video.
Perplexity, the AI chatbot you’ve probably never heard of (unless you read this newsletter), just added an interesting feature to their paid ($20/mo) plan. Image generation based on your chat! How does that work? Well, you start up a chat with Perplexity, then hit the image generation button and it creates something related to the topic you are discussing.
Aht Gallery
This week’s theme: Winter Wonderland. (Michelle Johnson via Midjourney, Dalle-E 3, Leonardo, Dream Studio.)
Unless otherwise noted, all images generated with Dall-E 3.