You know what you could do on this amazing day? Read Five Things and then become a paid subscriber! Awesome idea, right? ✌🏻 Oh, by the way, if you do not want to subscribe to Five Things, Five Things Tech and Five Things Running, you can select which one of the newsletters you want to read in your account settings. Five Things AI: Grok, Sakana, AI Literacy, AI Risk Repository, Advanced Voice ModeBefore you head into the weekend, read this!
Hallo and welcome to Five Things AI! This week we will take a look at how the image-generating capabilities of Grok lead to lots of hyperrealistic images of Trump, Musk, Biden, Taylor Swift and others. We also learn about an AI model that decides to modify its own code, which is just a tiny bit scary if you think about it. Then we get a better understanding about the thoughts and discussions in the scientific community about the usage of AI for research. We can also take a look at a database of 700 risks associated with AI and learn how to work with this when evaluating AI projects. Finally, let’s look at ChatGPT Advanced Voice Mode, which sounds really very advanced and also pretty weird. I know what you’re thinking. What are bizzare selection of articles. It’s not my fault that this all popped up this week. One more thing: If you haven’t already. subscribe. Thanks. 🕺 Musk’s AI chatbot sparks flood of bizarre generated images, from Micky Mouse saluting Hitler to Taylor Swift deepfakes
I’m not a bit surprised. Musk shows once again that he doesn’t care at all how the tools he’s providing could be misused. This obviously needs to be regulated as he is to careless to run an AI business. I assume that he likes the attention that X and xAI get. Research AI model unexpectedly modified its own code to extend runtime
This is both really amazing and scary. A AI model for scientific research modified its own code to better do the intended work. Wow. I’m sure that this will lead to interesting discussions among the developers. Quest for AI literacy
As AI becomes a part of our everyday lives, it becomes part of scientific research as well (as noted above…). This obviously poses tons of questions for scientists to get a better understanding of how this impacts their work and how to figure out the degree of the impact. How to use the AI Risk Repository
Researchers at MIT compiled a list of 700 risks associated with AI. This will be helpful in the months to come when AI gets deployed in more and more use cases. I Used ChatGPT's Advanced Voice Mode. It's Fun, and Just a Bit Creepy
“I’m using it most when I’m home alone and I want something to keep me company while researching articles and playing video games. The more time I spend talking with ChatGPT’s Advanced Voice Mode, the more I think OpenAI made a wise choice rolling out a less-flirty version than what was originally demoed. Don’t want to get too emotionally attached.” - I don’t know if I really need more voices at home, but I guess other people could enjoy having a companion. If you missed last week’s edition of Five Things AI, you can read it here:
That’s it for Five Things AI this week! 🤖 — Nico You're currently a free subscriber to Five Things. For the full experience, upgrade your subscription.
|
