security
Explanation
This single-panel comic shows a computer monitor displaying a CAPTCHA-style security prompt that reads: "Prove you are human by drawing a copyrighted Pokemon, using only penises." Below, the caption states: "There is a way to always block corporate AI. Companies are just afraid to use it."
The joke operates on several levels. First, it plays on the ongoing arms race between CAPTCHA systems (designed to distinguish humans from bots) and AI systems that have become increasingly capable of solving them. Traditional CAPTCHAs -- distorted text, image recognition puzzles, "click all the traffic lights" -- are being defeated by modern AI at near-human or better-than-human rates, making them less effective.
The comic proposes a CAPTCHA that would be unsolvable by any corporate AI for legal rather than technical reasons. A corporate AI could not complete this challenge because doing so would require it to generate copyrighted intellectual property (a Pokemon character, owned by Nintendo/The Pokemon Company) using sexually explicit content -- two things that corporate AI systems are specifically trained and constrained to refuse. Content filters block NSFW output, and copyright guardrails prevent reproduction of protected characters.
A human, by contrast, could easily draw a crude Pokemon made of penises with no legal or technical barrier -- the individual user is not bound by corporate content policies or IP licensing agreements.
The deeper joke is the observation that the things corporate AI cannot do are not defined by the limits of the underlying technology but by legal and corporate liability concerns. The phrase "companies are just afraid to use it" adds another layer: even though this CAPTCHA would be extremely effective, no company would deploy it because it would require their users to draw explicit content on their platform, creating its own legal and PR nightmare. So the "perfect" anti-AI security measure exists in theory but is unusable in practice for reasons of decency -- a Catch-22 of content moderation.