Skip to main content

THE THREATS OF ARTIFICIAL INTELLIGENCE - ANCAMAN KECERDASAN BUATAN

THE THREATS OF ARTIFICIAL INTELLIGENCE - 
ANCAMAN KECERDASAN BUATAN

Researchers at the Royal Institute of Technology KTH in Stockholm, Sweden, using recursive training AI combined with side-channel attacks have succeeded in breaking the CRYSTALS-Kyber public key encryption and key encapsulation mechanism recommended by NIST (National Institute of Standards and Technology) in July 2022 for post-quantum cryptography. 

  

The researchers added that “Deep learning-based side-channel attacks can overcome conventional countermeasures such as masking, shuffling, random delays insertion, constant-weight encoding, code polymorphism, and randomized clock”

It appears that artificial intelligence could become the most important threat to classical and post-quantum encryption algorithms and that protection against quantum attacks alone is no longer sufficient. Artificial intelligence attacks are becoming a major concern and must be addressed.

This does not mean that one should forget about cryptographic systems capable of resisting attacks by quantum computers, such as the SST© secure transmission system from PT SYDECO, which is not based on mathematical formulae and is therefore post quantum safe, but that one should also protect one's data and facilities against attacks by artificial intelligences.

The best protection for data in motion is to transport it via a VPN. The VPN is safe for the AI as long as the server is well protected because it is the server that the AI will attack.

Indeed, if the AI manages to compromise the VPN server, it can potentially access the data that is transmitted between the user and the server, even if this data is encrypted in the VPN tunnel.

By securing your computer network or simply your computer facilities with ARCHANGEL© NGFW, you are using your own VPN whose server is protected against all attacks because it is inserted in the ARCHANGEL defence system and is protected by 3 firewalls.

The best protection for data at rest is to protect access and limit it to authorised persons only.

But as the system data is not concentrated in one place, it is the whole system that must be protected and not just with a password, however sophisticated, but with a state-of-the-art system designed to protect against all types of attacks, providing protection in depth as ARCHANGEL© NGFW does.



Para peneliti di Royal Institute of Technology KTH di Stockholm, Swedia, menggunakan AI pelatihan rekursif yang dikombinasikan dengan serangan saluran samping telah berhasil memecahkan enkripsi kunci publik CRYSTALS-Kyber dan mekanisme enkapsulasi kunci yang direkomendasikan oleh NIST (Institut Standar dan Teknologi Nasional) di Juli 2022 untuk kriptografi pasca-kuantum.

Para peneliti menambahkan bahwa “Serangan saluran samping berbasis pembelajaran mendalam dapat mengatasi penanggulangan konvensional seperti masking, pengocokan, penyisipan penundaan acak, pengkodean bobot konstan, polimorfisme kode, dan jam acak”

Tampaknya kecerdasan buatan dapat menjadi ancaman paling penting bagi algoritme enkripsi klasik dan pasca-kuantum dan perlindungan terhadap serangan kuantum saja tidak lagi memadai. Serangan kecerdasan buatan menjadi perhatian utama dan harus ditangani.

Ini tidak berarti bahwa seseorang harus melupakan sistem kriptografi yang mampu menahan serangan oleh komputer kuantum, seperti sistem transmisi aman SST© dari PT SYDECO, yang tidak didasarkan pada rumus matematika dan oleh karena itu aman pasca kuantum, tetapi orang itu juga harus melindungi data dan fasilitas seseorang dari serangan kecerdasan buatan.

Perlindungan terbaik untuk data bergerak adalah dengan mengirimkannya melalui VPN. VPN aman untuk AI selama server terlindungi dengan baik karena itu adalah server yang akan diserang oleh AI.

Memang, jika AI berhasil mengkompromikan server VPN, AI berpotensi mengakses data yang dikirimkan antara pengguna dan server, bahkan jika data ini dienkripsi di terowongan VPN.

Dengan mengamankan jaringan komputer atau instalasi komputer Anda dengan ARCHANGEL© NGFW, Anda menggunakan VPN milik Anda sendiri yang servernya terlindungi dari segala serangan karena terintegrasi dengan sistem pertahanan ARCHANGEL dan dilindungi oleh 3 firewall.

Perlindungan terbaik untuk data saat istirahat adalah melindungi akses dan membatasinya hanya untuk orang yang berwenang.

Tetapi karena data sistem tidak terkonsentrasi di satu tempat, seluruh sistemlah yang harus dilindungi dan tidak hanya dengan kata sandi, betapapun canggihnya, tetapi dengan sistem canggih yang dirancang untuk melindungi dari semua jenis serangan. , memberikan perlindungan mendalam seperti yang dilakukan ARCHANGEL© NGFW.

#AI #VPN #Quantumcomputer #threats #Firewall #data #encryption #quantumsafe 

Comments

Popular posts from this blog

 Hospital Security in Question In a recent article titled "Cyberattacks: Public and Private Hospitals, Is the Worst Yet to Come?", Jean-Michel Tavernier1 provides a detailed analysis of why the medical sector is a prime target for hackers. He highlights the vulnerabilities that allow hackers to access sensitive data such as medical records, insurance information, and payment details. The compromise of this data can have severe consequences for individuals' privacy, financial security, and even personal safety, not to mention the financial damage to institutions and the risks to the quality of care provided to patients. System Vulnerabilities Tavernier points to the "excessive interdependence of the entire healthcare chain." Hospitals collaborate with a multitude of interconnected providers and partners, creating numerous opportunities for attackers. He suggests managing the attack surface (ASM), which means controlling and securing all entry points where unautho...
                                                                     ERRARE HUMANUM EST   The Inescapable Nature of Human Error and Its Implications in Cybersecurity To err is human; one could even say it is a defining characteristic of humanity. Who has never made a mistake, whether out of distraction, ignorance, or because it was provoked? No one is immune to making mistakes, and most of the time, they are forgivable, even if their consequences can be very damaging. However, the fundamental, unforgivable error is doing nothing to avoid situations that lead to mistakes. Thus, to minimize errors due to distraction, one should avoid multitasking (for example, a surgeon operating should not be distracted by a nurse recounting her latest adventures) and refrain from performing actio...

KEUANGAN & DATA PRIBADI PEMAIN GAME ONLINE DALAM BAHAYA

Melindungi Gamer Online: Memahami Risiko dan Solusi Dalam beberapa tahun terakhir, dunia game online telah mengalami pertumbuhan eksponensial, sejalan dengan meningkatnya nilai aset game. Namun, lonjakan popularitas ini juga membawa segudang risiko yang mengancam para pemain dan operator. Dari upaya peretasan dan pencurian akun hingga transaksi yang tidak sah dan eksploitasi data, bahaya yang mengintai di dunia digital selalu ada. Mengingat tantangan-tantangan ini, sangat penting untuk menjelaskan pentingnya langkah-langkah perlindungan yang kuat dan solusi inovatif. Memahami Lanskap Aset game, yang terdiri dari mata uang virtual, item, dan akun, adalah sumber kehidupan ekonomi game online. Nilainya melampaui ranah virtual, bahkan sering kali melampaui transaksi di dunia nyata. Meskipun demikian, perlindungan konsumen tradisional yang diberikan oleh layanan perbankan dan pembayaran tidak ada di ranah game. Operator platform game sering kali mengadopsi pendekatan laissez-faire, membuat ...