If you’re behind on what’s happening with the robot uprising, have no fear. Here’s a quick look at some of the weirdest and wildest artificial intelligence news from the past week. Also, don’t forget to check out our weekly AI write-up, which will go into more detail on this same topic.
Corporate America seems intent on automating art—a horrifying premise that I personally feel betrays the entire point of art and, thus, hope fails utterly and totally. But there’s good news: this week, a Washington D.C. judge ruled that AI-generated art should not be awarded copyright protections because no human was central to its creation. US District Judge Beryl Howell’s ruling importantly notes that “human authorship is a bedrock requirement” for any content that wants to be copyrighted. Copyright protections have been one of the hanging legal questions surrounding AI-generated content and this ruling sets a helpful precedent that says that art should only be considered art when a human was the prime mover behind its creation.
YouTube wants to help you make “music” with AI. The Google-owned video platform launched an AI music incubator this week that promises to “protect artists.” The new platform, which was created in partnership with music industry giant Universal Music Group (UMG), appears to be designed to protect well-known musical artists from having their content used in ways they don’t legally approve of. Licensing has been a big issue in the emergent trend of AI music, and this new partnership appears to be a stopgap in that ongoing legal debate.
In a broader sense, I’d just like to vaunt the argument that the term “AI-generated music” is something of a misnomer. Look, no offense, but if you think you’re making “music” by clicking a button and having an algorithm repackage sounds from actual music to create new synthetic auditory content that you put no actual effort into...you’re an idiot. That’s not music, it’s just noise.
Meta launched an automated translator this week that not only works in a multitude of language, but also helps translate for both speech and text. If that sounds impressive, there are some people who have concerns about what this new technology will do to people who actually want to be multi-lingual.
As Hollywood continues to ignore the reasonable demands of its own writers (you know, the people who actually make TV and movies worth watching), major studios have decided to dangle a big shiny carrot— a $1 million AI job—for any willing scab who wants to cross the picket line. Yes, studios have been posting more and more opportunities for “AI specialists.” As writers strike, partially to protest the use of this very technology in the creative process, the studios appear to be charging ahead with their plans to automate entertainment. There’s one thing I know with total certainty: if the studios get their way on this issue, there’s no way that movies and TV shows aren’t going to totally fucking suck for the foreseeable future.
Remember Microsoft Paint? No? Well, the company might shove AI into it. The company says it’s currently experimenting with integrating DALL-E like capabilities into the mundane scribbling tool.
Google’s AI search is wiling out. Yeah, the new-ish feature seems to be struggling, as our coverage found this week, the results for the tech giant’s AI-powered Search Generative Experience included stuff like exhortations on the “benefits of slavery” as well as a recipe for poisonous mushrooms. Thanks, Google!
Here’s something that’s definitely going to get a lot of use in the cybercriminal underworld: a new AI voice generator can allegedly fake your voice in up to 30 different languages. Voice-cloning schemes have already been used in heists in various parts of the world. As automated voice generation gets better and better, you can bet that cybercriminals are going to have a field day with this technology.