Buffout

This "kid level" argument fooled even Albert Einstein, Roosevelt, and many others. into developing atomic bomb. "If we don't build it Nazis will"

Using it was achieved by different kid level argument "It will prevent even more death and suffering"

Scientist have moral responsibility not to build weapons of mass destruction for example. Too bad even Einstein couldn't keep with your advice. Scientists are sometimes not so smart. They can be fooled like anybody else.

"Hey, maybe I can use it for starting 10 remote jobs", said software engineer.

- some guy uses ChatGPT to generate food recipe, doesn't check it. He poison himself.

- some guy uses ChatGPT to generate electronic circuit, doesn't check it. He electrocute himself.

So, what prevents chatGPT to use detection tool and fine-tune its response accordingly?

   > I have more success writing reusable libraries that do very, very fundamental stuff

Same here. Things like logical and mathematical operators. Memory access and stuff.

Never said it is easy to build simple reusable code. Code itself must be simple but process behind devising it may not be. If the code is not simple it is more prone to error.

"generalized" means the most simple code you can come up with. This simple code is then used thousands times from very different programs, and it never changes.

exactly my experience.

Authority is not always very good verifier. For example if you wan't to verify math behind answer, use something you trust, like calculator , abacus, ... whatever you trust.

Is this such a big problem? Everybody must verify data anyway. Otherwise bad things may happen. Like evil ai apocalypse that scientist have warned us about.

That is only half of the problem. Modern ransomware will steal data and also encrypt it. You may be able to restore data , but don't want data leak.

Researcher in article discovered that oven pings google , yandex and baidu to check if internet is present.

I think who ever coded oven firmware was fair because oven connectivity data will be equally shared with all major sides.

And also resilient, if it was only google, or only baidu or yandex, then they would have too much power to control the ovens. For example google may not respond and oven would never enter "internet present" state.

... Or they could just code different firmware for each region.

How will they contact us? We need Galactical mailbox. And you don't know how to build one unless you already have one.

BonziBuddy was spyware.

I have few apps that I Am the Only User, never shared them.

And the apps that I have shared, I never use them.

You should probably do opposite.

Interesting , article is about blog post, About how AI can be used to write Polymorphic malware. (writen also by AI).

What are they trying to prove?

That's the easy part. Checksum algorithms are public knowledge.

Looks like difficult part is inventing person with online history. Because if Facebook can't google the person it probably concludes person doesn't exist.

I mean how can you check government ID if it is legit? ( not how to photoshop ID. )

   > so they can verify that you didn't photoshop a fake government ID.

Huh. How?

Counter argument to system argument goes like this: If man memorizes all the rules and procedures he will be the system. Still doesn't know Chinese.

See my reply above, this are just bugs that are fixable.

"Timing attack" that will not be fixed is when police has warrant for your ISP, and police has ISP logs and destination server logs. (so it can compare the two)

No, this is not the timing attack I was thinking of. Your version of timing attack comes from inside tor network and that is solved by Entry Guards. And besides you can also be Tor relay and a client. This attacks are just bugs that are fixed.

The real timing attack that is not fixed and will not be (it is not in threat model), is when your ISP works with police (that has warrant) and gives them data. And police also controls server or exit node.

I don't think that is how timing analisis works. User can say to "entrance node" that it is also a node and that is how it can deny that it is originator. "Entrance node" doesn't know its position in a chain. Only Exit node knows its position.

So what if they were? Only exit node see the message (encrypted or not). And it doesn't know where it originated.

I speculate this is most likely case of a ISIS server run by FBI.

Article is about people on Russian forums trying to circumvent geofencing restrictions on ChatGPT.

This is strange, ( Taliban spokesperson using Twitter )

    Whenever Apple upgrades it's privacy scheme it seems the government becomes quite concerned requesting backdoors and such. However we have had apps like Signal and WhatsApp and a plethora of others with e2e for years.

link?

    Is it because our phones are compromised enough that they can access data before any user-space encryption ever takes place?

This is true. If your Android is compromised then e2e encryption is also broken, because you will find keys in memory.

There were no bugs then, there was very little electronics in a car. Not like in Uber.

Hi there , I'm coming from country where you could unlock many cars with you car key, yet car theft was rare.

Now I'm in this sci-fi future , 2023.