#nostr, you just zapped my post hard enough to pay for tasty tacos.

Amazing.

P.S. Current energy reconversion: 1 sat =18 kcal

nostr:nevent1qvzqqqqqqypzqcylrpk2qg7ktrq0uqv4wprj7k2ktj97rhqk8v25r7kfmy92f690qqstlpra2gkr22naz66uzty6p8z568ygk9akcwspssz4e8m2zsfj27gh58448

No notes.

#deepseek #openai

Things are sadly heading towards *all cars, all the time* and more and more car systems like infotainment break-ish when you try and disable these features.

Monday edition of *Car privacy is an absolute nightmare*:

Subaru's employee portal holds a year's worth of location data for all internet-connected cars.

We know this because it was vulnerable (now fixed). You could pull a year's worth of driving just with a license plate.

Props to Sam Curry & Shubham Shah for exposing it. Pic is a years' worth of Sam's mom's #Subaru locations.

I seriously doubt any owner has a clear idea that this data is being collected on them.

But the same thing is replicated for almost every car mfr (see the #Mozilla foundation report on car privacy link)

Literally no car owner has asked for their whip to be turned into a surveillance portal.

And yet..

Car companies feel basically no pressure to do right by customers, but experience a lot of incentives to mine their movements for money.

Sidenote: same (now closed) vulnerability also enabled remote unlocks & starts and a bunch of other highly undesirable things.

Reading list:

The Subaru research: https://samcurry.net/hacking-subaru

News report on it: https://www.wired.com/story/subaru-location-tracking-vulnerabilities/

Mozilla Foundation's key investigation into car privacy: https://foundation.mozilla.org/en/privacynotincluded/articles/its-official-cars-are-the-worst-product-category-we-have-ever-reviewed-for-privacy/

Who on #nostr is saying interesting things about #privacy?

Help me fill out my follows!

We are in the earliest days of sensing the privacy implications of incorporating #chatgpt / #deepseek et al. into our lives.

I think this period will be viewed as a time of a massive willing...but not fully witting... transfer of personal information, right down to our ways of reasoning & understanding the world...into the hands of a new set of companies & their CEOs.

The same companies that have the ambition to shape how we think.

And shape how society looks.

It shouldn’t take a panic over Chinese AI to remind people that most companies in the business set the terms for how they use your private data.

And when you use their AI apps, you’re doing work for them, not the other way around.

too kind! I call us researchers and ngos collectively an accountability ecosystem and I am proud to be part of it

I'm glad you found it interesting. Love the podcast.

thanks! I love making good trouble

Thank you. Excited to flex my freedoms

Thank you. I'm grateful to be able to do it. And have so many smart collaborators.

Thank you for the good energy!

Glad you enjoyed it! Which one?

Hi everyone, I'm JSR.

Me & my colleagues chase government hacking & censorship of dissidents & activists.

Heard of Pegasus spyware? Then you know about our work.

I'm part of the Citizen Lab, a ferociously independent research group based at the University of Toronto.

We try to 🥊punch above our weight.

Your device has probably gotten security updates that flowed from our collaborative investigations.

Craziest story? When we punked a team of mercenary🕵️ spies sent to target our research. https://www.youtube.com/watch?v=Z8IrU_jvnFk

I'm so proud of my colleagues & talented collaborators in the fight to serve fat helpings of good trouble & accountability.

All the good research I've worked on has been a team production... But everything I say here, especially the typos & bad ideas are100% mine.

This. No accident that legacy platforms slide towards restricting access by researchers.

After the acquisition of Twitter, an underreported play was freezing out any meaningful research into manipulation by gatekeeping the API & firehose.

So users are left with "this feels weird" & few ways to know more.

nostr:nevent1qvzqqqqqqypzqpxfzhdwlm3cx9l6wdzyft8w8y9gy607tqgtyfq7tekaxs7lhmxfqqst0hcjl3hdjj0udvjygvm6l3xqkzskdrtffpzgrjhmncfzv34gnjga5st0f

The official instinct to solve all problems by application of surveillance dystopia is a feature, not a bug.

Back in September, the FTC painted a picture of social media companies collecting a staggering amount of people's data.

Made it clear that platform self-regulation on privacy was a joke.

Respectfully, duh. But the question is: will any of that scrutiny continue?

Because the reckless disrespect of privacy & mining of interactions for profit is only going to accelerate.

For now that report is still online. We'll see for how long: https://www.ftc.gov/system/files/ftc_gov/pdf/Social-Media-6b-Report-9-11-2024.pdf

Here's what I found.

https://primal.net/e/note1r62wf5rn7gy3pre7g06c0swqf69yledrqudp8el3y9rn2z2c8kdsqhgy90

If you run a local version of an R1 derivative model it does not happen.

Although those models have interesting answers about Taiwan.

And if you try to spring the same question in a dialogue about something else, you get a bit more verbose a response.

Pay attention to the reasoning.