The idea is that generative AI will enable Samsung products to get a better understanding of how consumers use the products – for example, an oven recognizing what is being cooked in it or a fridge recognizing what ingredients are inside. This could allow appliances to understand users’ needs and respond accordingly.

“Understand users’ needs” being a euphemism for “spy on users’ habits and sell that info to advertisers.”

We’ve gone full circle: from having a manual for your new appliance, to having a LLM confidently make up some incorrect info about how to use your new appliance.

The paper states that they studied the HTML form element interactions but “not the keystrokes or content.”

There’s a big difference. Both are more invasive than we would like, but grabbing everything you type while in the app’s browser is much worse than measuring a true or false “did this person submit their comment or did they give up and leave it unsubmitted.”

Tiktok is getting the content of the text, which could be sensitive info, and it grabs from every site you visit, not just the social platform itself.

But I think the main issue is using the data for allegedly targeting of protestors and Chinese political opponents, more than the depth of the data collection itself.

TikTok has always been on the extreme end of tracking and surveilling its users. For example, research found that the app had the ability to record all keystrokes made by users in the in-app browser (i.e. keylogging). This kind of tracking is way beyond what other social media companies do and borders on malware.That’s one reason why the US, Canada, and others banned the use of TikTok on government devices.

A former TikTok employee also alleges in a sworn statement that TikTok stores its user data in China, that the CCP has full access to this data, and that the CCP used this data to spy on protestors in Hong Kong.

So their tracking goes way beyond what other companies do, and China uses that data for expressly political goals rather than simply selling ads to users.

When people claim that leaks “get people killed,” they’re referring to when undercover agents are identified while they’re in the field. The only secrets exposed in these leaks are the computer hacking techniques used by the US to spy remotely through compromised devices.

The so-called Vault 7 leak revealed how the CIA hacked Apple and Android smartphones in overseas spying operations, and efforts to turn internet-connected televisions into listening devices.

You could maybe say that closing off those surveillance channels prevented the CIA from learning about some attack, but that’s really tenuous. It also assumes that the CIA isn’t constantly developing new zero-day exploits so that they can continue to spy on just about everyone on the planet.

It’s always crazy to come into threads like these and see people say “I would murder as many elites as possible” without batting an eye, and in the same comment say “I could never give up hamburgers.” It’s some kind of insane self-soothing to throw all of the responsibility for a global issue onto a few scapegoats. It also shows that people have no intention of doing fuck all about climate change beyond typing up snarky comments on the internet.

People can misquote all kinds of studies they half remember to pretend that they have no responsibility for making changes, but that doesn’t make it true. Just as one example, first world countries’ per-capita rate of meat consumption alone is enough to push the world over our 1.5C warming target. But because it’s an inconvenience to make any changes to my life, I’m going to pretend I would personally kill scores of people rather than make a new recipe for dinner. We’re fucked

Not to mention that the existing Twitter infrastructure was already incredibly insecure before Musk even took over.

Twitter does little to monitor for so-called insider threats, employees or contractors who use their positions in the company to steal information, and instead leaves them “virtually unmonitored."

Twitter suffered security incidents significant enough to warrant a report to a government agency about once a week, with 20 breaches in 2020 alone.

Twitter devs can already take over user accounts since they all have prod admin access (which they need because Twitter still has no QA or staging environments). I can only imagine the potential for abuse once people’s finances get tied together with their account.