IP Lookup


What is AI-Slop?

Large-Language-Models (LLMs) are a type of artificial intelligence (AI) designed to generate and understand language / text. These models are often misued to generate outputs that appear correct without any human editing or manual intervention. Text outputs can subsequently be used for income generation via advertisement revenue and search engine optimization (SEO) poisoning, or by influencing public opinion with propaganda and synthetic political discourse using bot social media accounts. Vauge, low-effort prompting and a lack of specificity often leads these computer models to generate similar outputs that statistically average the patterns observed from their training data. LLM-slop can be identified by looking for some of the following writing patterns:

  1. Binary contrasts: "It's not X, it's Y."
  2. Throat-clearing openers: "Here's the thing" / "Let me be clear"
  3. Faux-insight: "What nobody tells you" / "The part that everyone misses"
  4. Colon reveals: "The best part: it learns."
  5. Dramatic fragments: "That's it. That's the whole thing."
  6. Superficial analysis: "...highlighting the team's commitment to innovation"
  7. Importance puffery: "...marks a pivotal moment" / "a testament to"
  8. Synonym cycling: "The agent handles your email. The assistant drafts replies..."
  9. Fake-profound endings: "The future isn't coming. It's already here."
  10. Em dash usage: "—"

Relevant XKCD

What is it?

This is a way to transfer your files directly from one device to another. To establish a connection to each other, you and the recipient exchange a set of generated "codes." After that, you can send the recipient as many files as you want (until you close the connection). Also note that there's a timeout while connecting, so type fast!

Relevant XKCD

Why?

In peer-to-peer file transfer, your data gets sent directly to the recipient and doesn't have to travel through an intermediary server, so it's faster and more private. You also don't have to worry about the file size limits that exist in emails or texts.

How?

This implementation uses WebRTC to establish a connection between browsers and is (almost) completely serverless. In order for peers to discover each other, a STUN (Session Traversal Utilities for NAT) server must be used. This server helps to find a network path between the two peers as a part of the ICE (Interactive Connectivity Establishment) protocol. The peer initiating the connection generates an SDP (Session Description Protocol) offer, which must be sent to the other peer. After the recipient receives the first peer's offer, they generate a SDP answer, which must be sent back to the first peer. After both peers complete the handshake, the connection can be established and data can start passing directly between them. Normally, a signaling server is used to send the initial handshake. However, I wanted to rely on as little external servers as possible (without hosting my own instance), so I decided to let the users send this data directly to each other. To make this feasible, the handshake is heavily compressed and boilerplate data is reconstructed client-side. For additional ease-of-use, the "short-code" generated is an even shorter key linked to handshake data via a 3rd party paste site. Please note, if the initial ICE connection fails, it's most likely an issue with a firewall or symmetrical NAT, so sending data directly between the two peers isn't possible.

Share Files





Receive Files