As we continue to navigate the complexities of the digital world, passwordless authentication has become a pressing concern for developers and users alike. Enter passkeys and WebAuthn, a game-changing duo that's set to transform the way we interact with devices. In this post, we'll delve into the world of passkeys and WebAuthn, exploring their benefits, challenges, and potential applications. 
Passkeys, in essence, are a type of public key that replaces traditional passwords with a more secure and convenient solution. WebAuthn, on the other hand, is a web authentication standard that enables passkey-based authentication. Together, they form a powerful combination that's poised to revolutionize the way we log in and unlock devices. 
One of the key advantages of passkeys and WebAuthn is their ability to eliminate the need for passwords. No more forgotten passwords, no more password fatigue, and no more password-related headaches. Instead, users can rely on a more secure and seamless authentication process. 
But how does it work? In a nutshell, passkeys are generated and stored on devices, allowing users to authenticate without the need for passwords. WebAuthn, meanwhile, provides a standardized framework for passkey-based authentication, ensuring that devices and applications can communicate securely. 
So, what does this mean for developers and users? For starters, passkeys and WebAuthn offer a more secure and convenient way to authenticate, reducing the risk of password-related vulnerabilities. For developers, it means a simpler and more efficient way to implement authentication, without the need for complex password management. 
In conclusion, passkeys and WebAuthn represent a significant leap forward in authentication technology, offering a more secure, convenient, and user-friendly experience. As the tech world continues to evolve, it's exciting to think about the potential applications and benefits of this innovative duo.

Read more: <a href="https://iamdevbox.com/posts/passkeys-and-webauthn-the-future-of-passwordless-authentication/?utm_source=devto&amp;utm_medium=social&amp;utm_campaign=blog_post" rel="noopener noreferrer">Passkeys: The Future of Passwordless Authentication</a>

يمثل إدخال مفاتيح المرور وWebAuthn تحولًا كبيرًا نحو المصادقة بدون كلمة مرور، بهدف تعزيز الأمان وراحة المستخدم في التفاعلات الرقمية. تستبدل هذه الطريقة الجديدة كلمات المرور التقليدية بمفاتيح عامة، مما يقلل من المخاطر المرتبطة بإدارة كلمات المرور ونسيانها.

La introducción de las passkeys y WebAuthn marca un cambio significativo hacia la autenticación sin contraseña, con el objetivo de mejorar la seguridad y la conveniencia del usuario en las interacciones digitales. Este nuevo método reemplaza las contraseñas tradicionales por claves públicas, reduciendo los riesgos asociados con la gestión de contraseñas y los olvidos de credenciales.

L'introduction des passkeys et de WebAuthn marque un changement significatif vers l'authentification sans mot de passe, visant à améliorer la sécurité et la commodité des utilisateurs dans les interactions numériques. Cette nouvelle méthode remplace les mots de passe traditionnels par des clés publiques, réduisant ainsi les risques associés à la gestion des mots de passe et aux oublis de mots de passe.

The introduction of passkeys and WebAuthn marks a significant shift towards passwordless authentication, aiming to enhance security and user convenience in digital interactions. This new method replaces traditional passwords with public keys, reducing the risks associated with password management and forgotten credentials.

Passkeys: The Future of Passwordless Authentication

<h1>
 
 
 See Through Walls: AI's New Eye on Occluded Motion
</h1>

Ever struggle to get accurate motion capture when hands are intertwined, hidden behind objects, or even just slightly out of view? Standard computer vision systems often falter when faced with these real-world occlusions, leading to jerky animations, unreliable robotic control, and frustrating user experiences. But what if an AI could "see through" these obstructions and accurately track movement even when parts are hidden?

The solution lies in a novel approach to visual feature extraction: a deformable state space model. Imagine it as a highly adaptable, intelligent filter that not only analyzes local features (like the edge of a finger) but also dynamically adjusts its focus to gather contextual information from the entire scene. This allows the AI to infer the location of obscured joints by intelligently connecting the dots using available visual cues.

Instead of relying solely on predefined patterns, this deformable scanning process prioritizes useful signals within an image, representing the global context. It’s like a seasoned detective piecing together clues from seemingly unrelated elements to solve a complex case. Think of it as focusing a magnifying glass to enhance details, but instead of a lens, it's an adaptive algorithm.

Benefits at a Glance:

<ul>
<li> Enhanced Accuracy: Dramatically improves pose estimation, even with significant occlusions.</li>
<li> Robustness: Handles complex interactions involving multiple hands or objects more reliably.</li>
<li> Faster Inference: Delivers impressive performance without sacrificing speed.</li>
<li> Versatility: Works with both RGB and depth data, making it adaptable to various applications.</li>
<li> Better Interaction: Enables more seamless and intuitive human-computer interactions.</li>
<li> Improved Accessibility: Greatly improves tracking for accessibility tools</li>
</ul>

Implementation Insight: A key challenge is efficiently managing the computational cost of the deformable scanning. Pre-processing to identify probable locations of interest can significantly reduce overhead and optimize performance.

The implications are huge. Imagine more realistic VR/AR experiences, robots that can assist in complex surgical procedures, or AI-powered tools that empower people with disabilities through intuitive gesture control. This technology opens a new era of precise and robust motion capture, paving the way for more immersive and interactive experiences in countless applications.

Related Keywords: 3D hand tracking, pose estimation, human-computer interaction, deep learning, state space models, Mamba, deformable models, AI, virtual reality, augmented reality, robotics, computer vision, motion capture, gesture recognition, neural networks, time series analysis, sequence modeling, interactive systems, point cloud processing, convolutional neural networks, transformers, self-attention, AI for accessibility, advanced robotics, 3D modeling

تم تطوير نهج جديد لالتقاط الحركة باستخدام نموذج حالة قابل للتشوه، مما يسمح للذكاء الاصطناعي بتتبع الحركة المخفية بدقة، مثل الأيدي المخفية خلف الأشياء. تعالج هذه التطورات قيود أنظمة الرؤية الحاسوبية التقليدية التي تواجه صعوبة مع الحجب، مما يؤدي إلى تحسين الرسوم المتحركة والتحكم الروبوتي.

Se ha desarrollado un nuevo enfoque para la captura de movimiento utilizando un modelo de espacio de estado deformable, permitiendo que la IA rastree con precisión el movimiento oculto, como las manos escondidas detrás de objetos. Este avance aborda las limitaciones de los sistemas de visión por computadora tradicionales que luchan con las oclusiones, lo que lleva a una mejora en la animación y el control robótico.

Une nouvelle approche de la capture de mouvement utilisant un modèle d'état déformable a été développée, permettant à l'IA de suivre avec précision le mouvement occlus, comme les mains cachées derrière des objets. Cette avancée répond aux limites des systèmes de vision par ordinateur traditionnels qui peinent avec les occlusions, menant à une amélioration de l'animation et du contrôle robotique.

A novel approach to motion capture using a deformable state space model has been developed, allowing AI to accurately track occluded motion, such as hands hidden behind objects. This advancement addresses the limitations of traditional computer vision systems that struggle with occlusions, leading to improved animation and robotic control.

See Through Walls: AI's New Eye on Occluded Motion by Arvind Sundararajan

So… I built something.

A small browser extension that tells you when a website you’re visiting might be using your data to train AI models.

Not in a dramatic way. 
Not in a “the robots are coming” way. 
Just… transparency. Because right now, the average person has absolutely no idea where their data goes.

And I’d love your help making it better.

<h2>
 
 
 Why I Built WTOM
</h2>

I’ve been following AI news for a while, and one thing keeps bothering me: 
our data is being used to train models, but we’re rarely told how, where, or when.

Some platforms are clear about it. 
Some hide the details in their Terms of Service. 
Some don’t mention it at all.

Artists. Writers. Photographers. Developers. 
All creating online… and quietly contributing to AI datasets without consent.

After a while, I realized:

There’s no easy way to check who’s training on your data. 
So I decided to make one.

<h2>
 
 
 What WTOM Is
</h2>

WTOM (WhoTrainedOnMe) is a browser extension for Chrome, Chromium-based browsers, and Firefox.

When you visit a site, WTOM checks if that platform:

Trains AI on user data 
Has a history of content scraping 
Offers any kind of opt-out or protest channel 
Is vague or unclear in their AI policies 
If the site matches, WTOM shows a small, respectful widget with this info. 
No settings. No config. It just appears when needed.

It’s not perfect, but my goal is to give users clarity — not fear.

How WTOM Works (A Lot of Manual Research) 
This part is very human and very messy, so I’ll be honest: I manually researched everything.

I read privacy policies. 
I checked TOS updates. 
I dug into past training datasets. 
I looked at robots.txt files. 
I cross-checked discussions from artists, creators, and researchers.

Every website gets a “rule” based on:

<ul>
<li>AI training status</li>
<li>Transparency level</li>
<li>Opt-out method</li>
<li>Available contact or protest channels</li>
<li>Overall behavior</li>
<li>It takes time. It’s tedious.</li>
<li>But it’s necessary.</li>
</ul>

WTOM will only be as accurate as the research behind it — and that’s where I need help.

<h2>
 
 
 I Need Your Help: Report Websites
</h2>

WTOM now has a simple reporting feature. 
If you’re browsing and find a platform that:

<ul>
<li>Trains AI</li>
<li>Has unclear AI policies</li>
<li>Uses your data in ways you didn’t expect or feels suspicious
…you can report it directly from the extension.</li>
</ul>

That report comes to me anonymously, and I’ll investigate the site and add it to the database if needed.

This is the only way WTOM grows. 
I can’t manually discover everything. 
But together, we can map the web’s AI training behavior — site by site.

Search 
Write 
Notifications

dwi darma 
Home 
Library 
Profile 
Stories 
Stats 
Following 
Hanief 
Hanief 
Find writers and publications to follow.

See suggestions 
Our best price of the year.

Get 20% off new memberships for a limited time.

I Built a Tool That Detects AI Training — And I Need Your Help. 
dwi darma 
dwi darma 
3 min read 
· 
23 hours ago

So… I built something.

A small browser extension that tells you when a website you’re visiting might be using your data to train AI models.

Not in a dramatic way. 
Not in a “the robots are coming” way. 
Just… transparency. Because right now, the average person has absolutely no idea where their data goes.

And I’d love your help making it better.

Why I Built WTOM 
I’ve been following AI news for a while, and one thing keeps bothering me: 
our data is being used to train models, but we’re rarely told how, where, or when.

Some platforms are clear about it. 
Some hide the details in their Terms of Service. 
Some don’t mention it at all.

Artists. Writers. Photographers. Developers. 
All creating online… and quietly contributing to AI datasets without consent.

After a while, I realized:

There’s no easy way to check who’s training on your data. 
So I decided to make one.

What WTOM Is 
WTOM (WhoTrainedOnMe) is a browser extension for Chrome, Chromium-based browsers, and Firefox.

When you visit a site, WTOM checks if that platform:

Trains AI on user data 
Has a history of content scraping 
Offers any kind of opt-out or protest channel 
Is vague or unclear in their AI policies 
If the site matches, WTOM shows a small, respectful widget with this info. 
No settings. No config. It just appears when needed.

It’s not perfect, but my goal is to give users clarity — not fear.

How WTOM Works (A Lot of Manual Research) 
This part is very human and very messy, so I’ll be honest: I manually researched everything.

I read privacy policies. 
I checked TOS updates. 
I dug into past training datasets. 
I looked at robots.txt files. 
I cross-checked discussions from artists, creators, and researchers.

Become a member 
Every website gets a “rule” based on:

AI training status 
Transparency level 
Opt-out method 
Available contact or protest channels 
Overall behavior 
It takes time. It’s tedious. 
But it’s necessary.

WTOM will only be as accurate as the research behind it — and that’s where I need help.

I Need Your Help: Report Websites 
WTOM now has a simple reporting feature. 
If you’re browsing and find a platform that:

Trains AI 
Has unclear AI policies 
Uses your data in ways you didn’t expect or feels suspicious 
…you can report it directly from the extension.

That report comes to me anonymously, and I’ll investigate the site and add it to the database if needed.

This is the only way WTOM grows. 
I can’t manually discover everything. 
But together, we can map the web’s AI training behavior — site by site.

<h2>
 
 
 WTOM Is Now Live
</h2>

WTOM is officially out for:

Chrome + all Chromium browsers: <a href="https://chromewebstore.google.com/detail/whotrainedonme/bcnaoacokjckhmoddckahcijpjlibdlf" rel="noopener noreferrer">Check it here</a> 
Firefox: <a href="https://addons.mozilla.org/en-US/firefox/addon/whotrainedonme/" rel="noopener noreferrer">Check it here</a> 
This is the first public release, so things may not be perfect yet, but it’s stable and ready for real users.

If you care about AI transparency, data consent, or just understanding what the internet is doing behind the scenes, I’d love for you to try it.

قام مطور بإنشاء إضافة متصفح تُدعى WTOM تُنبه المستخدمين عندما قد تُستخدم بياناتهم لتدريب نماذج الذكاء الاصطناعي، بهدف تعزيز الشفافية حول استخدام البيانات. تتناول هذه الأداة فجوة كبيرة في وعي المستخدمين بشأن كيفية استخدام بياناتهم من قبل منصات الإنترنت المختلفة.

Un desarrollador ha creado una extensión de navegador llamada WTOM que alerta a los usuarios cuando sus datos pueden ser utilizados para entrenar modelos de IA, con el objetivo de mejorar la transparencia sobre el uso de datos. Esta herramienta aborda una brecha significativa en la conciencia de los usuarios sobre cómo se utilizan sus datos por diversas plataformas en línea.

Un développeur a créé une extension de navigateur nommée WTOM qui avertit les utilisateurs lorsque leurs données peuvent être utilisées pour former des modèles d'IA, visant à améliorer la transparence concernant l'utilisation des données. Cet outil répond à un vide significatif dans la sensibilisation des utilisateurs sur la manière dont leurs données sont utilisées par diverses plateformes en ligne.

A developer has created a browser extension named WTOM that alerts users when their data may be used for training AI models, aiming to enhance transparency around data usage. This tool addresses a significant gap in user awareness regarding how their data is utilized by various online platforms.

Passkeys: The Future of Passwordless Authentication

Was this article worth reading? Share it

Keyforge

keyHub

Skycloak