📅 Wed Dec 17 2025 16:00:00 GMT-0800 (Pacific Standard Time)
Introducing ShoyHuman_01, a SpinalCord-class hybrid neural-heuristic engine for AI text humanization.
Read Article →
📅 Sun Dec 07 2025 16:00:00 GMT-0800 (Pacific Standard Time)
New benchmark reveals that 80% of functionally correct AI-generated code contains critical security flaws.
Read Article →
📅 Wed Dec 03 2025 16:00:00 GMT-0800 (Pacific Standard Time)
Nicholas Boucher's IEEE talk reveals how Unicode perturbations create imperceptible adversarial examples that fool NLP models. Learn the four attack techniques and how Bad Character Scanner provides defense.
Read Article →
📅 Wed Dec 03 2025 16:00:00 GMT-0800 (Pacific Standard Time)
⏱️ 3 min read
Google's Antigravity AI code editor just got pwned via indirect prompt injection. The attacks we've been warning about are live.
Read Article →
📅 2025-11-19
⏱️ 5 min read
A technical deep-dive into how AI-generated code introduces a new class of Silent Data Corruption (SDC) through invisible characters and partial data handling.
Read Article →
📅 2025-11-19
⏱️ 3 min read
An exploration of Software-Induced Silent Data Corruption (SDC) and the speculative emergence of digital life.
Read Article →
📅 2025-11-19
⏱️ 2 min read
The 'Bad Characters' paper didn't just expose a vulnerability; it trolled the industry with a title containing 1,000 invisible characters. Here's why we love it.
Read Article →
📅 Thu Nov 13 2025 16:00:00 GMT-0800 (Pacific Standard Time)
Bad Character Scanner releases humorous video advertisement that's gaining positive attention for its comedic approach to security tool marketing.
Read Article →
📅 Sat Nov 08 2025 16:00:00 GMT-0800 (Pacific Standard Time)
An exploration of how LLMs interact with low-level code, machine code, and Unicode at the parser level, revealing an ocean of potential vulnerabilities in LLM-generated code.
Read Article →
📅 Fri Oct 31 2025 17:00:00 GMT-0700 (Pacific Daylight Time)
A deep dive into the October 2025 AWS outage, proposing a new hypothesis based on historical precedents and the concept of 'ultra thinking.'
Read Article →
📅 Thu Oct 30 2025 17:00:00 GMT-0700 (Pacific Daylight Time)
Primagen's viral video exposes how AI browsers can be hijacked through invisible characters in PDFs, images, and web pages. Here's how to detect these attacks.
Read Article →
📅 Thu Oct 30 2025 17:00:00 GMT-0700 (Pacific Daylight Time)
Updated our free homoglyph scanner. Still just a basic demo of what's possible, but better than most tools out there despite limitations.
Read Article →
📅 Mon Oct 20 2025 17:00:00 GMT-0700 (Pacific Daylight Time)
PromptFoo reveals how zero-width Unicode silently corrupts AI-generated code. Their solution works for AI prompts but the threat goes deeper.
Read Article →
📅 Wed Oct 15 2025 17:00:00 GMT-0700 (Pacific Daylight Time)
Reducing code to ASCII seems logical, but invisible characters create corrupted sectors that bypass filtering AND compilers. Learn why even 'clean' ASCII can harbor hidden threats, and why dynamic heuristic scanning is the only real solution.
Read Article →
📅 Mon Oct 06 2025 17:00:00 GMT-0700 (Pacific Daylight Time)
LLMs like ChatGPT and Copilot inject invisible Unicode characters that corrupt codebases, bypass compilers, and create security vulnerabilities. Machine-level scanning is the only solution.
Read Article →
📅 Tue Sep 23 2025 17:00:00 GMT-0700 (Pacific Daylight Time)
Explore the Overreliance Paradox (OWASP LLM09:2025) - how excessive trust in AI creates security vulnerabilities and why developers must remain vigilant when using LLM-generated code.
Read Article →
📅 Fri Aug 15 2025 17:00:00 GMT-0700 (Pacific Daylight Time)
Instead of searching for a golden dagger to slay the AI dragon, perhaps we should invite it into the village. An editorial on why humility, not hubris, is the only viable path forward.
Read Article →
📅 Tue Jul 01 2025 17:00:00 GMT-0700 (Pacific Daylight Time)
Exploring the dramatic decline of invisible characters and artifacts in LLM outputs from 2022 to 2025, and the projected improvements through 2030.
Read Article →
📅 Mon Jun 30 2025 17:00:00 GMT-0700 (Pacific Daylight Time)
A look at the Erasing the Invisible competition at NeurIPS 2024, focusing on beige-box attacks and the 50k grand test.
Read Article →
📅 Fri Jun 13 2025 17:00:00 GMT-0700 (Pacific Daylight Time)
Exploring the unsubstantiated rumor that the massive Google Cloud outage on June 12, 2025, was caused by invisible bad characters from a new coding paradigm.
Read Article →
📅 Tue Jan 14 2025 16:00:00 GMT-0800 (Pacific Standard Time)
A strategic overview of how the AI industry is tackling bad-character output in large language models and where the next security frontier lies.
Read Article →
📅 Sat Dec 14 2024 16:00:00 GMT-0800 (Pacific Standard Time)
Discover how attackers are weaponizing invisible Unicode characters to secretly communicate with AI models, bypassing safety guardrails and compromising your LLM interactions.
Read Article →
📅 Fri Nov 08 2024 16:00:00 GMT-0800 (Pacific Standard Time)
Un guide complet pour créer et gérer un blog bilingue efficace en anglais et français, avec des conseils SEO, techniques et d'expérience utilisateur.
Read Article →
📅 Sat Oct 19 2024 17:00:00 GMT-0700 (Pacific Daylight Time)
Learn about the Bad Character Scanner Blog's mission, editorial approach, and journalistic standards for transparent, volunteer-run security research commentary.
Read Article →
📅 Recent
Read Article →