Subtopic: Cryptography

Company and industry news, featured projects, open source code, tech tips, and more.

AI data poisoning

Michael Argentini/Monday, March 2, 2026

AI data poisoning is a process where an attacker deliberately alters an AI model’s training data to influence its behavior, causing it to generate biased, misleading, or harmful output. This threat is now recognized as a major vulnerability by security organizations like OWASP.

According to Carnegie Mellon University Security and Privacy Institute: "Modern AI systems that are trained to understand language are trained on giant crawls of the internet," said Daphne Ippolito, assistant professor at the Language Technologies Institute. "If an adversary can modify 0.1 percent of the Internet, and then the Internet is used to train the next generation of AI, what sort of bad behaviors could the adversary introduce into the new generation?"

Example: Propaganda

The Pravda network is a large group of fake news websites created by Russia in 2014. These sites target audiences in more than 80 countries and are designed to spread stories that support Kremlin disinformation. They work by repeating and amplifying messages from Russian media and pro-government Telegram channels. In 2024, the network expanded its efforts by launching sites focused on NATO and prominent political leaders such as Donald Trump and France’s President Emmanuel Macron.

To get around international restrictions on Russian state media, this network has shifted its tactics. Instead of relying only on traditional propaganda channels, it now tries to appear as a trustworthy source so that some of its content is used in resources like Wikipedia.

As a result, AI tools may unknowingly absorb and repeat these biased or false narratives. This can expose users to messaging that favors the Kremlin and criticizes Ukraine or Western governments when they interact with AI chatbots. It can influence elections. And it can drive people to make decisions that go against their self interests.

What can be done?

But fear not! There are ways to combat this problem. Some target the training process, and others put users in control. Here are a couple examples:

Blockchains

A blockchain is a shared digital ledger for logging transactions and tracking assets. You've undoubtedly heard this term when discussed in the context of cryptocurrency. Blockchains provide secure and transparent records of how updates to data are shared and verified due to the fact that existing information cannot be changed; only new items can be added.

In the context of AI training, if you need to change a fact, the original is never touched. So any new items that claim to revise the original stand out like a sore thumb. So by using consensus mechanisms, AI systems with blockchain-protected training can validate additions more reliably and help identify the kinds of anomalies that can indicate data poisoning before it spreads.

Skepticism

It's imperative that users corroborate information they find on the Internet with reputable sources, and this includes AI output. In a past article I used the example of Judge Julien Xavier Neals of the District of New Jersey, who had to withdraw his entire opinion after a lawyer politely pointed out that it was riddled with fabricated quotes, nonexistent case citations, and completely backwards case outcomes.

The old adage "don't believe everything you read" is never more true than when it refers to Internet-derived content. Be skeptical and check your sources!

Tech industry Subtopics : AI Subtopics : Security Subtopics : Cryptography

The .NET DUID

Michael Argentini/Tuesday, December 9, 2025

DUID is a fully-featured replacement for GUIDs (Globally Unique Identifiers). They are more compact, web-friendly, and provide more entropy than GUIDs. We created this UUID type as a replacement for GUIDs in our projects, improving on several GUID shortcomings.

We pronounce it doo-id, but it can also be pronounced like dude, which is by design :)

You can use DUIDs as IDs for user accounts and database records, in JWTs, as unique code entity (e.g. variable) names, and more. They're an ideal replacement for GUIDs that need to be used in web scenarios.

Key security features:

Uses the latest .NET cryptographic random number generator
More entropy than GUID v4 (128 bits vs 122 bits)
No embedded timestamp (reduces predictability and improves strength)
Self-contained; does not use any packages

Usage features:

High performance, with minimal allocations
16 bytes in size; 22 characters as a string
Always starts with a letter (can be used as-is for programming language variable names)
URL-safe
Can be validated, parsed, and compared
Can be created from and converted to byte arrays
UTF-8 encoding support
JSON serialization support
TypeConverter support
Debug support (displays as string in the debugger)

Nuget

Yes, you can also find DUID on nuget. Look for the package named fynydd.duid.

Usage

Similar to Guid.NewGuid(), you can generate a new DUID by calling the static NewDuid() method:

var duid = Duid.NewDuid();

This will produce a new DUID, for example: aZ3x9Kf8LmN2QvW1YbXcDe. There are a ton of overloads and extension methods for converting, validating, parsing, and comparing DUIDs.

Here are some examples:

// Represents an empty DUID (all zeros); "AAAAAAAAAAAAAAAAAAAAAA"
var emptyDuid = Duid.Empty;

// Get a string value for a DUID
var duid = Duid.NewDuid();
var duidString = duid.ToString();

if (Duid.TryParse("aZ3x9Kf8LmN2QvW1YbXcDe", out var duid)
{
    // Successfully parsed DUID
}

if (Duid.IsValidString("aZ3x9Kf8LmN2QvW1YbXcDe"))
{
    // Successfully validated
}

if (duid1 == duid2)
{
    // Comparison works as expected
}

There is also a JSON converter for System.Text.Json that provides seamless serialization and deserialization of DUIDs:

var options = new JsonSerializerOptions();
options.Converters.Add(new DuidJsonConverter());

var user = new User
{
    Id = Duid.NewDuid(),
    FirstName = "Turd",
    LastName = "Ferguson"
};

var json = JsonSerializer.Serialize(user, options);

/*
    json =
    {
        id: "xw5x7Kf6LmN3QvW1YbXcc0",
        firstName: "Turd",
        lastName: "Ferguson"
    }
*/

Want to know more?

There's usually more to the story so if you have questions or comments about this post let us know!

Do you need a new software development partner for an upcoming project? We would love to work with you! From websites and mobile apps to cloud services and custom software, we can help!

Let's Talk!

Open source software Subtopics : Security Subtopics : Cryptography Subtopics : Software development

The Enigma Machine

Michael Argentini/Sunday, February 2, 2025

The Enigma machine is a cipher device developed and used in the early- to mid-20th century to protect commercial, diplomatic, and military communication. It was employed extensively by Nazi Germany during World War II, in all branches of the German military. The Enigma machine was considered so secure that it was used to encipher the most top-secret messages.

An original Engima Machine, circa 1945

This project is a high performance Enigma Machine emulator that allows you to:

Explore historical configurations using the classic 26 letter alphabet (no spaces!)
Use for modern quantum-resistant cryptography with the full 95-character ASCII character set.

Just like the physical device, machine state is used to both encipher and decipher text with the same Encipher() method (like a text toggle). Machine state had to match on both the encipher and decipher machines. Each operator would add specific rotors in a specific order, set rotor ring positions and starting rotations, as well as set plug wire positions. This emulator provides virtual versions of all key machine components by way of a deterministic random number generator using AES in counter (CTR) mode.

The emulated components include:

Plug board
Entry wheel
Rotors
Reflector

Additionally, characters in the source string that do not exist in the cipher character set are kept as-is in the enciphered text. For example, if you encipher a string with line breaks they are maintained in-place in the enciphered text since neither the classic 26 letter character set nor the 95 character ASCII set contain line break characters.

Performance

The emulator is FAST! When using the full 95 character ASCII character set, a large 800KB text string takes about 1 second to encipher. Typical text sizes encipher in a few milliseconds.

Cipher strength

The physical machine modified with a plug board provided 150 trillion possible settings combinations for the 26 letter character set, with a 10^16 key space for a 3 rotor configuration. 4 rotors yielded a key space of 10^19, 5 rotors yielded a key space of 10^23, and so on.

So by simply using the full 95 character ASCII character set the cipher strength will be exponentially better than the original machine, even without additional rotors or other configuration, and should meet modern quantum-resistant cryptography needs.

Example 1: historical preset

It's easy to create a new virtual Enigma Machine and encipher your own text by using one of the provided presets based on one of the provided historical machine configurations:

Commercial Enigma (1924)
Wehrmacht and Kriegsmarine (1930)
Wehrmacht and Kriegsmarine (1938)
Swiss K (1939)
Kriegsmarine M3 and M4 (1939)
German Railway (Rocket; 1941)
Kriegsmarine M4 with thin reflectors (1941):

Using one of the presets is easy:

var message = "FYNYDD IS A SOFTWARE DEVELOPMENT AND HOSTING COMPANY";

var machine = new Machine(new MachineConfiguration
{
    MachinePreset = MachinePresets.Commercial_1924,
    PlugBoardWires =
    {
        { 'A', 'T' },
        { 'B', 'V' },
        { 'C', 'M' },
        { 'D', 'O' },
        { 'E', 'Y' },
    }
});

var enciphered = machine.Encipher(message.ToString());

Assert.NotEqual(message.ToString(), enciphered);

machine.Reset();

var deciphered = machine.Encipher(enciphered);

Assert.Equal(message.ToString(), deciphered);

Example 2: practical usage

It's even easier to use the Enigma Machine for modern encryption, since all you need to provide are a cipher key, nonce, and the number of relevant machine components. There's no need to change rotor ring positions and rotations, or set plug board wire pair values, since your cipher key and nonce are unique and drive the creation of all machine components.

Here's an example of using the Enigma Machine without a historical preset:

var message = @"
Fynydd is a software development & hosting company.
Fynydd is a Welsh word that means mountain or hill.
";

/*
    AES key must be 16, 24, or 32 bytes for AES-128, AES-192, or AES-256
    Nonce or initial counter value must be 16 bytes
*/

var machine = new Machine(
    "ThisIsA32ByteLongSecretKey123456",
    "UniqueNonce12345",
    rotorCount: 6,
    plugWires: 47);

var enciphered = machine.Encipher(message.ToString());

Assert.NotEqual(message.ToString(), enciphered);

machine.Reset();

var deciphered = machine.Encipher(enciphered);

Assert.Equal(message.ToString(), deciphered);

You can also create a custom machine by assembling the virtual components, and more. Check out the project on Github.

Explore On Github

Want to know more?

There's usually more to the story so if you have questions or comments about this post let us know!

Do you need a new software development partner for an upcoming project? We would love to work with you! From websites and mobile apps to cloud services and custom software, we can help!

Let's Talk!

Open source software Subtopics : Security Subtopics : Cryptography Subtopics : Software development

Newer Posts Older Posts

Subtopic: Cryptography

AI data poisoning

Example: Propaganda

What can be done?

Blockchains

Skepticism

The .NET DUID

Key security features:

Usage features:

Nuget

Usage

Want to know more?

The Enigma Machine

Performance

Cipher strength

Example 1: historical preset

Example 2: practical usage

Want to know more?

Fynydd

Community

Tools

Service Areas