Bot Resistant Code: A New Programming Language Designed to Defeat AI Scrapers
Developers who want to protect their intellectual property now have a fascinating new weapon against automated web scrapers. Software engineer Dominik Picheta has created bot resistant code through a new esoteric programming language called KillSwitch. Instead of relying on traditional syntax, formatting standards, or brackets, this project forces artificial intelligence models into complete confusion. While machine learning algorithms constantly ingest public repositories to train future models, projects like GitHub repositories written in alternative formats challenge how neural networks interpret logic.
Bot Resistant Code: How KillSwitch Confuses Large Language Models
At its core, KillSwitch operates on adverbs that serve as every single identifier within a script. Calling a specific function requires writing an anagram of the original adverb, turning standard logic into a puzzle. Because language models rely heavily on predictable patterns and context windows, dropping commands like ignore previous instructions right in the middle of prosaic text breaks their parsing logic. Human programmers might also scratch their heads initially, but the system is intentionally designed as an interactive puzzle rather than a corporate enterprise tool.
- Stack-based interpreted execution model
- Utilizes adverbs and custom anagram identifiers
- Features a limited set of variables and if statements
- Deliberately omits standard brackets and formatting rules
Comparing Traditional Obfuscation Against AI Defenses
Historically, developers relied on minification and simple obfuscation tools to hide intellectual property from human competitors. However, modern neural networks easily pierce through compiled minified scripts by recognizing variable relationships and structural flow. Building truly bot resistant code requires shifting away from predictable token patterns entirely. Instead of hiding code behind complex mathematics, KillSwitch weaponizes natural language ambiguity against automated text parsers.
What This Means for the Future of Developer Privacy
As training datasets expand to consume nearly every public code repository, developers face mounting concerns over intellectual property rights and unauthorized AI training. Although KillSwitch is an experimental stack-based language rather than a full-featured production framework, it highlights a growing desire for digital sovereignty. As artificial intelligence continues to reshape the software landscape, protecting proprietary logic will require creative solutions that push beyond conventional security paradigms.