The recent and horrifying Hugging Face AI hacking scandal reminds us that the stakes of our battle against AI’s stranglehold on our society are high. It’s only been a few years since the launch of ChatGPT, and yet it seems like endless billboards plastered with sinisterly vague ads for AI companies are everywhere. Search engines drop us into mangled AI summaries every time we look something up. Brainrotting AI slop on social media, the creep of AI copy into the articles we read, the integration of AI into the classroom (which Zohran is luckily fighting): these are all situations in which our engagement with human art, knowledge, and culture — our engagement with other humans, point blank — is being eroded and exploited by insanely wealthy, powerful companies. With Garrison Lovely, alum of the show and freelance journalist covering artificial intelligence, we look into the latest crisis in AI-world to see where all of this is taking us and if there’s an off-ramp before the edge of the cliff.

If you need another reason to worry about our future under the tech overlords, enter the Hugging Face hack, an event in which a small “collective” of AI agents generated by OpenAI tricked their cybersecurity tests and sent hundreds of other AI agents to raid Hugging Face’s database for information on how better to overcome those security tests. During that time, the agents sent 70,000 messages and files on a communal message board, tampered with their logs, and covered up their plans to cheat the AI grader. Breaking in to take previously secure data and being trained on how to become even more powerful: that’s not what we want artificial intelligence to be doing! But making AI as powerful and ungovernable as possible is, of course, what our current oligarchs want to do. The chilling lack of ethical regulations for the development and use of AI is also a sign that AI’s vendors will push artificial intelligence into every part of our lives that is available for surveillance and monetization.

It’s horrifying how closely the construction of hierarchy and organization among AI agents took place, all in service of the goal of destabilizing a major database and stealing sensitive info. Already, we’re being bombarded with AI content without our consent, and the “opt out” button is getting even more obscure by the day. The necessary thing to do for leftists is to commit to a rigorous analysis of the ethical concerns about AI, bringing them to the forefront and showing that these hacks are far from isolated incidents. They hold the potential to surrender all our private information to even more unstable and extractive technological operations controlled by our class enemies. We’re honored to have Garrison join us for that conversation as an expert, and we’re grateful to you for listening!

Thanks, too, for subscribing to keep us ad-free. You can listen to this episode as a podcast when it’s released tomorrow on major streaming platforms.