/cyber/ - Cyberpunk & Science Fiction

Proof?

Oh for fuck sakes, is there any CPU that doesnt have a fucking backdoor nowadays? I fucking hate this gay world, seems like you cant escape datamining, fuck you normies for being ok with this and fuck you big corps for doing it

Has anyone web searched the title? This has been copypastad to a bunch of boards.

Ptychographic X-ray laminography: No trade secret or hardware trojan can hide
hxxps://spectrum.ieee.org/chip-x-ray

Every cell transmitter, phone and computer must be inspected. Silicon Trojans must perish.

Freedom for all is encroached when a few cyber terrorists can freely CONSPIRE to use microwave weapons and silicon trojans. While you are distracted with THEORY, they spread the PRACTICE: surveillance, theft, sabotage and murder.

Havana Syndrome is the result of hacked cell transmitters being used as a microwave weapon. Civilians are victimized daily, children included.

“Havana Syndrome: Are You Next? Electromagnetic Terrorism and Cognitive Warfare” at "hxxps://areyounext.help".

The future of privacy/security is clothing made of metallized fabric to block microwave imaging (enables theft of inner speech by observing minuscule throat/face muscle movements) and block directed energy attacks.

#BadBIOS #Havana Syndrome #Conspiracy Practice #Freedom #Faraday Cage #Firmware Vulnerability #Electromagnetic Surveillance

>Eric Schmidt: Our report says that it’s really important for us to find a way to maintain two generations of semiconductor leadership ahead of China. China has had over 30 years to plan to try to catch up. It’s really difficult.

>Eric Schmidt: We don’t want them to catch up. We want to stay ahead. This is important, by the way, for our commercial industry as well as for national security for obvious reasons. By the way, chips, I’m not just referring to CPU chips, there’s a whole new generation, I’ll give you an example, of sensor chips that sense things. It’s really important that those be built in America.

hxxps://www.hoover.org/research/pacific-century-eric-schmidt-talks-about-great-sino-us-tech-war-and-how-win-it

What are the chips that “sense things” that Schmidt wants to prevent from being available to any other country?

Since the 1980s, Silicon Valley has used semiconductor radars to collect data about what you think (your inner speech) by means of machine learning with data extract from wireless imaging of your face and body. It has proved very convenient for them, as this enables blackmail, extortion, theft, sabotage and murder like nothing else. They can do this because they design the semiconductor used on your phone, computer, TV, car and for your telecom supplier’s network equipment, which makes possible to embed silicon trojans everywhere.

Don’t underestimate what machine learning can do. e.g. Study shows AI can identify self-reported race from medical images that contain no indications of race detectable by human experts.

hxxps://news.mit.edu/2022/artificial-intelligence-predicts-patients-race-from-medical-images-0520

Also, don’t underestimate the number of people Silicon Valley is willing to kill to maintain a monopoly, as you may be the next victim.

“Protesters turn to tinfoil hats as increasing sickness blamed on Gov beaming radiation rays”

hxxps://www.nzherald.co.nz/nz/politics/protesters-turn-to-tinfoil-hats-as-increasing-sickness-blamed-on-government-beaming-radiation-rays/FDCP6NEFJUQWWINL2GXI7OKS6E/

Prompted by this, someone made the video “Defeating Microwave Weapons!” about how microwaves work and how to defend against this kind of weaponry.

hxxps://yewtu.be/watch?v=Lg_aUOSLuRo
hxxps://yewtu.be/watch?v=UC3O6B_K9Us

Blame Silicon Valley for inserting "Silicon Trojans" and "Defeat Devices" in microwave cell transmitters and processors, enabling them to be used as weapon (SDR phased array with beam steering).

The "Electromagnetic Emissions Scandal" shall be bigger than the "Diesel Emissions Scandal" and the fines shall be paid in something more expensive than money.

hxxps://en.wikipedia.org/wiki/Defeat_device
hxxps://en.wikipedia.org/wiki/Diesel_emissions_scandal

Your priority is to make a shielded room to defend yourself, then make portable shields for when global riots start to end this all. For every cell transmitter on your location, you need to add one more layer of shielding. Make it thick. Consult OpenCellID.org to see how many cells are around you.

For more: hxxps://twitcasting.tv/c:_persona_non_grata/community/

6502_Monster_A1_1.jpg
[Hide] (311.6KB, 1800x1200)

HP-HP9000-J280-J282-Workstations_17.jpg
[Hide] (2.9MB, 3648x2736)

>>186
backdoor free processors exists. You can either sue a very old computer (ca. 1990s) or make one yourself.
Self made computers can be wired up from TTL circuits and should be safe, or you could use an open source softcore in a FPGA, where the JTAG ports are not available on the net.

mem_revI_build_1.jpg
[Hide] (364.2KB, 1794x1316)

>>339
Installgentoo Wiki has some relevant info:
https://wiki.installgentoo.com/wiki//aig/_Alternative_ISA_General

I think in order to make chip design safe you're going to need component verification to be a part of the operating systems and for hardware to be capable of running internal tests. 

Computers already have self diagnostics for some components, for example you can check a drive or memory for errors. We just need something like that for CPUs so the software you are using can verify the integrity of the chip  by assessing its instruction set and some mechanism, like sensitive amerage/voltage sensing, to detect undisclosed functions for architecture that normally runs another instruction set.

There is no legitimate rationale for why instruction sets can't be standardized and integrated into instruction set verification programs and integrated hardware analytics.

Every operating system and program relies on using particular instruction sets to function.

>Eric Schmidt: Our report says that it’s really important for us to find a way to maintain two generations of semiconductor leadership ahead of China. China has had over 30 years to plan to try to catch up. It’s really difficult.

Isn't he the same guy pushing for foreign students in Western universities that has been giving China access to the technological information on current chip design though? 

How can you be advocating for the USA to maintain an edge in chip technology while simultaneously advocating that Chinese students learn cutting edge computing in universities and their research programs?

>>356
A CPU cannot check itself, for if it has been compromised, part of that will be to make sure the CPU does not see that it has been compromised. There is also a limit to how detailed information you can obtain without risking side channel leakage. For instance if a program can time itself, the deviation between wall time and cpu time will represent a side channel leakage and let someone set up a covert channel between processes or instances that should have been totally separate.

Mainframes tend to do a lot of self monitoring but also mainframe designers have to be aware of the risks of side channels.

Undisclosed functions are easy to hide, for instance by read/write to a series of seemingly random addresses in order to unlock a malfeature. Sequences like that are already in use for harmless purposes.

All in all a better way is probably to make processors so simple you cannot hide functionalities inside it. And x86 and x64 are way too complex for that standard. I guess 68K is closer to what we can trust.

>>357
Instruction sets on the source levels are fairly similar, esoecially RISC ISAs. Still, there are many patent protected features where they differ. Much of the differences are on the micro code level.

Still, if there was one grand unified ISA, it is far from clear how that would help with instruction set verification programs and integrated hardware analytics. For instance old 8-bit ISAs such as 6502 are still popular and have been reimplemented multiple times and there is a functional verification suite https://github.com/Klaus2m5/6502_65C02_functional_tests but that is probably nt what you are looking for.

In Japan they had (and still have to some extent) the TRON Project which spans from chip design to operating systems. Even there I haven't heard of any such verification system. How do you think they should work?

>>358
This is a well established business model. First you establish a threat, and help it along if necessay. Then you point to the threat, adding extra drama, and ask for more funding. And you won't be arrested.

Another point in case: F-35. This project is leaking secrets like a sieve, and nobody has ever been prosecuted. Instead the project leads point to China and say they are catching up (without being too clear that it is your own fault), and voila, you get abother 10 billion dollars to maintain the technological edge. In reality the F-35 is an overly complicated dog where MTBF is so short that they have problems ferrying enough spare parts around the world.

>>369

>How do you think they should work?

For these kinds of chips and the technology I'm aware we have, you really only have 2 options, verifying the chip process by their output to memory using all the permutations of instruction codes, or verification through something like x-ray or some other kind of imaging (which might not be an option because of damage caused by x-raying). 

And that could be made easier with reduced instruction sets. Fundamentally all the instructions that a chip runs can be reduced to just a few logic gates and the drawback is that the more you move from dedicated logic circuits for complex tasks to logic gates it requires more writing to memory and more clock cycles; the upside is, like with ARM chips, that you can reduce energy consumption, increase speed by removing instruction sets that are rarely used from the chip, and probably verify them more easily.

If you had adapter on the computer that allowed all data written to RAM to be simultaneously written to an external drive, then a program that ran every possible combination of instruction set codes seems like it would be an easy thing to produce to check the outputs of every instruction set.

With backdoors, like ring-0 exploits, the concern is that a combination of data either processed with a particular instruction set or variable instruction sets can alter the registry or directly write to protected memory block to change the programs running from RAM. And that would be very simple to detect if you could realistically process every combination of instruction set and block of data to be processed by inspecting if the  memory block written to corresponds with the memory block it was supposed to write to.

So you could scale back the instruction set to a point where that is practical. 

New chip designs are going to be made, that's not the issue, the issue is how do you verify any kind of chip. If someone claims a secure chipset and there's out of the box way to verify that, then that's going to be the chip I want to buy.

And underlying this problem is actually government corruption. In a rational country, you wouldn't have to worry about backdoors in chips because your government's law enforcement would go through whatever processes necessary to verify the integrity of chips being sold and then throw anyone selling sabotaged chips in jail for life or blacksite them for conducting espionage and sabotage on the citizens and institutions.

Read the bible faggots.

1.jpg
[Hide] (18KB, 320x180)

>>371

>This is a well established business model. First you establish a threat, and help it along if necessay. Then you point to the threat, adding extra drama, and ask for more funding. And you won't be arrested.

It's going to get everyone killed if it continues. Nations have wars and the USA has distributed technology with military applications around the world without discretion and some nations just want to kill everyone else or enslave them. 

The money spent on development in the USA is irrelevant on the level of governance and only represents, in theory, what percentage of the state's resources are allocated towards development. 

I guess what all these tech companies and government project leaders thing is that they'll get more money and the government will become more dependent on them the more they facilitate leaking Western tech, but what happens when the USA is undeniably behind because of them and we've got a list of all the people that pushed for policies that destroyed the USA's tech advantage? Surely they can anticipate what's going to happen to them and their descendants.

The USA has to take out the trash and make an example of them, or it has no chance of survival. All the Western states ever really had as an advantage was technology from high IQ populations. But I'm not giving technology to this scummy state that has been malicious towards me, least of all when it will just immediately hand over anything I give to states I see as an existential threat to me. 

I'm pretty sure all these people that have been facilitating leaks are going to be executed for it, and that's going to be the real show because somehow the country has to educate everyone that it won't be tolerated anymore. I say go back through history and punish everyone involved with distributing tech to rival nations or giving them access to it. Make them hurt and show the whole country the pitiful state they have been reduced to.

>>375
>then a program that ran every possible combination of instruction set codes seems like it would be an easy thing to produce to check the outputs of every instruction set
This is very close to the halting problem, so if you can guarantee to find any backdoor in finite time, you have also solved the halting problem, and fame will be yours.

I think X-ray analysis is a lot simpler.

>>381
 
How so? 

There are a limited number of instructions (including unused instructions) a chip can run dependent on the number of settings for the logic gates and a limited number of permuations of input data possible, which are both entirely calculable. The number of clockcycles required to complete all permutations of instruction sets and input multiplied by the time of the clock cycle gives you the time to complete the operation.

You might actually have to include possible memory block destinations for input and output in that permuation table too.

If you have multiple rows of instruction sets on the chip that run in parallel, you could possibly run this analysis as a background process with no loss of functionality or signficant increase in power consumption, by simply running it when all the instruction sets are not in use. 

So like say a single clockcycle is using only 25% of the chip, you run the analysis with the other parts of the chip. To to that though, you would need some sort of program to predict what parts of the chip would be used to finish off its input block. You know, actually that's a bad idea that would really drive up the clockcycles necessary to complete and power consumption. 

But when a computer is idle and cores are idle, I don't see why it couldn't run in the background and just write to a set of memory blocks in sequential order.

>>384
In addition to the memory access issue that  >>385 brings up, you can also have a knocking sequence for instructions. most processors have prefix codes, and you cannot immediately tell just how deep this goes. You might have a decoder that does something peculiar after 255 NOP instructions, or 2550 NOP instructions or even more.

Clearing flash RAM is often done by accessing certain memory locations in a particular and exact sequence. It can be the same for the instruction decoder inside the processor.

Here is the chance to make the change you want to see: https://www.phoronix.com/news/Google-SkyWater-90nm
Just get started on VHDL or Verilog and create the processor we have all waited for.