OpenAI’s policy changed quietly to remove the ban on using its tools and models for military and weapon development purposes.
Author: Abhimanyu
Unwrapping the fast-evolving AI popular culture.
OpenAI Pleads It Can’t Work Without Copyrighted Material
OpenAI says it’s impossible to work with just public domain work, and the company will be doomed if it doesn’t train its models using copyrighted work.
Google Writes “Robot Constitution” to Make Sure Its Robots Don’t Kill Us
DeepMind writes a Robot Constitution based on Isaac Asimov’s Three Laws of Robotics to ensure safety when the LLMs inside Google’s bots work on tasks involving humans, animals, sharp objects, electrical appliances, etc.
MasterKey, an AI that Jailbreaks Other AI LLMs
A team of researchers developed an AI tool specifically trained to jailbreak other AI LLMs like the ones powering ChatGPT and Bard, allowing it to bypass content filters and safety protocols with a 20%+ success rate.
Leak Reveals Artists’ Work Midjourney Used for Training
Artists including Cy Twombly, Damien Hirst, Andy Warhol, Yayoi Kusama, and others have been used to train Midjourney, the AI-based image generator, raising new concerns over copyrights and privacy.
Microsoft Quietly Launches Copilot App with Free GPT-4 & DALL-E 3 Access
Microsoft launched the Copilot app with free GPT-4 and DALL-E 3 access (for original image creation) over the holidays on the Play Store and the App Store without any official announcement.
4 LLMs Face Off in Chess, GPT-4 Beats Gemini Pro and Mixtral 8x7B
GPT-4, GPT-3.5 Turbo, Gemini Pro, and Mixtral 8x7B were programmed to play chess against each other in an experiment by a Redditor. GPT-4 won by a large margin.
AI Can Find Locations in Your Photos
Three Stanford graduates trained OpenAI’s CLIP on 500K Google Street View images and it guessed locations from pictures better than a GeoGuessr veteran.
Study Reveals ChatGPT Cheats Under Pressure
Researchers found that ChatGPT would almost always use a means, no matter if forbidden, if it has access to it, under fabricated stress or pressure.
NY Times Sues OpenAI for Copyright Infringement
The Times joins the fight against OpenAI’s LLMs using publishers’ work without permission and publicly featurng it to its users with a fresh lawsuit.