Jen King, PhD

Very honored to receive the inaugural UC Berkeley Tech Integrity Award today for academic research from my alma mater. citrispolicylab.org/berkeleytechpo…

@alex_prompter Thanks @alex_prompter for highlighting our research!

🚨 Holy shit… Stanford just exposed that every major AI company is using your private conversations to train their models by default. They analyzed the privacy policies of OpenAI, Google, Meta, Anthropic, Microsoft, and Amazon. Reviewed 28 separate documents across all 6 companies. The findings are worrisome. Every prompt you type. Every file you upload. Every personal detail you share. All of it feeds directly into model training the moment you hit send. That health question you asked ChatGPT at 2am? Training data. Legal situation you described to Claude? Training data. The photo you uploaded to Gemini? Training data. Some companies retain your conversations INDEFINITELY. Amazon, Meta, and OpenAI have no confirmed deletion timeline for certain chat data. Your most private conversations could sit on their servers forever. It gets worse for kids. Four out of six companies allow children aged 13-18 to use their chatbots, and most don’t treat children’s data any differently. Kids’ conversations are likely getting fed into model training by default. Kids who can’t legally consent to it. Here’s something most people missed: enterprise customers are opted OUT of training by default. You, the consumer paying $20/month? Opted IN. Companies paying thousands? Protected automatically. There’s a two-tiered privacy system and you’re on the wrong side of it. OpenAI even frames the opt-in with guilt. Their settings page says “Improve the model for everyone.” Stanford’s researchers flagged this as a textbook dark pattern designed to make you feel bad for protecting your own data. Meta’s contractors told reporters they routinely see identifiable personal information in the chat data they review. Journalists were able to positively identify at least one real person from chat transcripts shared with them. The privacy policies themselves? Stanford had to dig through 6 separate documents just for OpenAI alone. Most real disclosures were buried in sub-policies no normal person would ever find. The researchers said it was challenging for THEM to piece it together. For consumers? “Practically impossible.” Only Microsoft explicitly stated they try to remove personal data like names, phone numbers, and addresses before training. The rest are either vague about it or completely silent.

The AI boom poses new challenges for privacy. What kinds of risks do we face and what can we do about it? @StanfordHAI fellow @kingjen describes the main findings from a recent white paper: stanford.io/3TCJJ55

@MarietjeSchaake Thank you @MarietjeSchaake ! You gave an exemplary talk today.

Congratulations to @kingjen on being awarded the Tech Integrity Award! ↘️

@nrmarda @mozilla Congrats Nik! 🥳🎉

Excited to be joining @Mozilla as their Technical Lead for AI Governance! Come read about how we're making AI safer and better, and share your thoughts about what we should do next:

@sorch1 Sorry. Welcome to parenthood. It's . . . a thing.

New paper alert! @acmeinhardt and I lay out a set of provocations regarding the impact of existing data privacy regulations in the US and EU on artificial intelligence. Our motivating question: can we have both data privacy and AI?

Bay area @Apple stores are experiencing a rash of thefts. But I find it impossible to believe that the floor models at Apple stores have no form of bricking or tracking mechanism. The company makes AirTags, after all! Or do they consider floor models disposable?

@privacypen Did you also have a pet dragon?

that's my girl! this is 3 mins from my childhood home. she's so pretty when she's mad.

@alexhanna This is a company that in the very early days used to require prospects to report their SAT scores!

Google loves to emphasize how smart its employees are. They even have a section in onboarding about how it's common to get impostor syndrome! Dear reader, there's no reason to get impostor syndrome.

RED ALERT: Buried in the House intelligence committee’s Section 702 “reform” bill, which is schedule for a floor vote as soon as tomorrow, is the biggest expansion of surveillance inside the United States since the Patriot Act. 1/11

@WillRinehart They don't have to be. @PamelaSamuelson's work on this issue remains both relevant and prescient: jstor.org/stable/1229511…

I don't want people to own their data, per se, I want them to be able to control it, says @kingjen. We can use various kinds of technologies to control our data. #DataGovGenAI2023 But control rights are property rights!

Packed house today for the Data Governance in the Age of Generative AI event. I will be in and out. The event details are here: datagovhub.elliott.gwu.edu/calendar_event… As I hear interesting comments or citations, I will be tweeting them out in this thread. #DataGovGenAI2023

I'm excited to share this press release from GW Law School announcing our new GW Center for Law & Technology. Through the Center, we're building out our privacy and tech curriculum and activities, including a new JD concentration in privacy/tech and more law.gwu.edu/gw-law-launche…

This article👇was the first I'd heard of using one-off GPS tracking to pursue stolen vehicles (vs. pursuit, which several bay area cities prohibit due to the risk of harm to the public). It requires decent aim and presence at the scene, but better from a surveillance perspective?

I was just turned away from an announced press conference with elected @AlamedaCountyDA Pamela Price. Every other member of the media who came was let in without any checking of credentials. @FACoalition @rcfp @SPJ_NorCal

Defaults matter. 👇👍 nytimes.com/2023/11/20/opi…

Hey, that's us! 😁 Have y'all tried Permission Slip yet? Circle Y or N

👇👍👍👍

Debate over the Kids Online Safety Act (KOSA) has been driven by emotion without a path forward. @‌KingJen and I examine KOSA and suggest how to overcome the stalemate for @‌techpolicypress. Here’s the bottom line. 🧵 1/9 techpolicy.press/overcoming-fea…