mouse_

35 ms·

271.
by mouse_ 1y ago
Good take
272.
by mouse_ 1y ago
the only security paradigm that is 100% foolproof is to assume breach. Taking away any number of users freedoms, big or small, does not change that. Old computers, before sandboxing and Windows defender and real-time protection, were more s
273.
by mouse_ 1y ago
Chronically online accelerationist death cultists find increasingly abstract ways to call people "NPCs" Somehow, not The Onion
274.
by mouse_ 1y ago
isn't the whole selling point of ML the idea that operators no longer need to be as skilled? it feels like the goal posts have been moving as of late.
275.
by mouse_ 1y ago
Nerds got taken aside and talked to about how it's not nice or cool to brag about IQ score so they invented a new artificial metric to brag about.
276.
by mouse_ 1y ago
> Extraordinarily talented people are doing work that LLMs already do better, out of spite. So what, people should just stop doing any tasks that LLMs do subjectively better?
277.
by mouse_ 1y ago
ironically I'm not allowed to view the comments on this thread because Twitter doesn't work the way it used to generally speaking though, I believe we're witnessing a digital enclosure movement.
278.
by mouse_ 1y ago
In my opinion, it is C. It puts utility first and has barely any features at all, yet still enough to have become the basis for the modern computing ecosystem.
279.
by mouse_ 1y ago
Dude, just install the closed source screen reader that tries to find PII/credit cards/social security numbers on your computer. It's for Cyber Security!
280.
by mouse_ 1y ago
I find it extremely challenging to believe you actually think that. I can't even host my own email anymore without all of my messages getting filtered as spam, because if your email service doesn't play nice with Google, then you
281.
by mouse_ 1y ago
> According to academics, one risk scenario involves collaboration between AI bots. Just imagine: hundreds of AI-driven social media profiles begin to pop up online, weaving narratives about certain companies. The information spread isn’
282.
by mouse_ 1y ago
When billions of dollars are involved, never attribute to stupidity what could be adequately explained by malice.
283.
by mouse_ 1y ago
One less thing to disable and hide on every new installation is a win.
284.
by mouse_ 1y ago
The only way to ensure your bitcoin is squeaky clean is to mine it yourself.
285.
by mouse_ 1y ago
He should make a startup sound for Haiku OS
286.
by mouse_ 1y ago
More choice is good.
287.
by mouse_ 1y ago
> When LLMs were still nascent and we didn't fully trust their output Why is this in past tense? Is the article trying to imply it solved the confabulation problem or what? Get real.
288.
by mouse_ 1y ago
I was gonna say nginx, lol. It's smaller and more portable than python afaik
289.
by mouse_ 1y ago
SDL is cool!!!
290.
by mouse_ 1y ago
This reeks of "Defiant Jazz"
291.
by mouse_ 1y ago
"Could" is a very charged word in a sentence like that.
292.
by mouse_ 1y ago
The end goal is to fire 100% of the workforce to secure an infinite, automated, exponential money printer. AI is merely a justification that he anticipates will sound nice enough to early VCs he's afraid will dump on him.
293.
by mouse_ 1y ago
I wonder how many healthy (>=~70% capacity) miles the batteries in these cheap EVs will do. Hopefully it's at least 100k, in order to compete with combustion engines?
294.
by mouse_ 1y ago
Too bad they look bad on standard DPI monitors (Still common!).
295.
by mouse_ 1y ago
> that doesn't mean you're going to get it. Totally wrong; as a computer user, I can very confidently say that computer users are going to get what they want. They may get what they want by switching to amateur, insecure softwa
296.
by mouse_ 1y ago
Great idea
297.
by mouse_ 1y ago
Happy to see this, but, Firefox for Android can do full flavored UBO, no?
298.
by mouse_ 1y ago
Related: https://en.m.wikipedia.org/wiki/Betteridge%27s_law_of_headli...
299.
by mouse_ 1y ago
Remember --- Assume Breach!
300.
by mouse_ 1y ago
Sounds like a threat

More ›