How's everyone's Friday going so far? Time to make it better, weirder, and/or cooler with the weekly roundup. On the podcast this week: how a couple of Harvard students cobbled together a pair of Meta's smart glasses that automatically dox people just by looking at them, and how data and AI are factoring into the Hurricane Helene aftermath. In the subscribers-only section, we discuss a hacked dataset that shows the prompts used on an AI companion site.
This segment is a paid ad. If you’re interested in advertising, let's talk.
Your personal data can easily be found with a simple Google search—DeleteMe is here to help. Join a growing community of privacy champions, from individuals to forward-thinking businesses, seeking to stop the leak at the source. We're offering 404 Media readers 20% off all consumer plans.
Listen to the weekly podcast on Apple Podcasts, Spotify, or watch it on YouTube. Unsplash WORDPRESSED As of this week, WordPress.org users are now required to agree that they are not affiliated with website hosting platform WP Engine before logging in. It’s the latest shot fired by WordPress co-creator Matt Mullenweg in his crusade against the website hosting platform. The checkbox on the login page for WordPress.org asks users to confirm, “I am not affiliated with WP Engine in any way, financially or otherwise.” Users who don’t check that box can’t log in or register a new account. As of Tuesday, that checkbox didn’t exist. I talked to community members about why this crosses a new line. Collage via 404 Media BOTS HIRING BOTSA job auto-applier is one of the most popular new projects on Github right now, and is actively being used by thousands of people to automatically apply for jobs en masse on LinkedIn. Jason wrote about how this is a response to “feeling like applying to and interviewing for jobs can be an endless process, which are often filled with requirements to complete tedious assignments and go through many rounds of interviews spread out over the course of months.” Unsplash AT ALL COSTSA group of Wikipedia editors formed WikiProject AI Cleanup with the goal to protect one of the world’s largest repositories of information from the same kind of misleading AI-generated information that has plagued Google search results, books sold on Amazon, and academic journals.“A few of us had noticed the prevalence of unnatural writing that showed clear signs of being AI-generated, and we managed to replicate similar ‘styles’ using ChatGPT,” Ilyas Lebleu, a founding member of WikiProject AI Cleanup, told Emanuel. “Discovering some common AI catchphrases allowed us to quickly spot some of the most egregious examples of generated articles, which we quickly wanted to formalize into an organized project to compile our findings and techniques.” Collage via 404 Media / Screenshots via Muah.ai UNCENSORED AND HACKEDA hacker targeted a website that lets users create their own “uncensored” AI-powered sexual partners and stolen a massive database of users’ interactions with their chatbots.The data, taken from a site called Muah.ai and viewed by 404 Media, includes chatbot prompts that reveal users’ sexual fantasies. Joseph wrote about how in many instances, users are trying to create chatbots that roleplay child sexual abuse scenarios. These prompts are in turn linked to email addresses, many of which appear to be personal accounts with users’ real names. AND NOW, A WORD FROM A FRIEND...Hi there, my name is Ryan Broderick and I write Garbage Day, an award-winning internet culture newsletter. And I’m throwing a big event all about the election this month. It’s called “America, The Final Season,” and it’s live at the Bell House in Brooklyn on October 23rd. Unlike other election events, mine will be fun and not depressing. You can get your tickets here! But also I am extending a very special offer to 404 Media readers. The first five people that email me at ryan@garbageday.email with the subject line “Gimme tickets plz” will get a free ticket to the show. (RIP my inbox...) MORE GOOD POSTSReplying to ‘I Applied to 2,843 Roles’: The Rise of AI-Powered Job Application Bots, L.K. wrote: I think there’s a high likelihood of this turning into a really shitty arms race if these automation tools take off. Employers will have to increase tedious requirements to thwart bots, the bots will get more sophisticated, and so on. The end result will make the job application process substantially worse and create higher barriers to entry. Maybe if we’re lucky, we’ll eventually come all the way back around to visiting a place in person and dropping off a resume to skip all the bullshit.
There are a ton of great comments and conversations happening on that story; supporters get commenting privileges so get in there! And to Hurricane Helene and the ‘Fuck It’ Era of AI-Generated Slop, RenderClouds said, I started paying for 404 because I truly dont think any other tech outlet is writing hard enough hitting articles about AI. What the hell is the point of running a tech news site if the people who actually understand and pay attention to tech are too afraid to actually criticize its effect on society? Thank you for your unrelenting coverage. Its so despicable to turn people away from having to look at images of real people suffering (and therefore, real people they can tangibly help with names and houses). This trend makes me horribly sad, but at least some people are still diligent about it.
BEHIND THE BLOGThis is Behind the Blog, where we share our behind-the-scenes thoughts about how a few of our top stories of the week came together. This week, we discuss a sensitive data breach, the terrible process of sorting and sending job applications, and how Wikipedia is battling AI slop. JOSEPH: Usually when I publish an article revealing a data breach, I don’t get all that many emails from readers. I imagine they read it and move on. There is a particular type of breach though where I often get inundated with emails: those that touch on sex or adultery. That definitely applies to the piece I published this week called Hacked ‘AI Girlfriend’ Data Shows Prompts Describing Child Sexual Abuse (obviously this headline isn’t explicitly about sex or adultery, it’s about something much, much more concerning, but it does also discuss those themes, or at least it touches on sex). After that article, I’ve received more emails and texts than usual from people asking me to give them the data, or perform lookups for certain people inside. One came from someone who claimed their boyfriend was one of the people whose info was included in the hack. “I wanted to ask if you know if there's any way I can access his info specifically? I don't want to know what others were searching on the website, so I don't need the full file. I just wanted to confirm if my boyfriend was using it for nefarious purposes. Do you have any advice on how I can get this info?” Read the rest of Joseph's Behind the Blog, as well as Sam, Jason, and Emanuel's, by becoming a paid subscriber.
|