Airparser

Automorphic offers a range of language model solutions, such as infusing knowledge into models using only 10 samples, self-improving capabilities, rapid loading/stacking of fine-tuned adapters, OpenAI API compatibility, and the Automorphic Hub for publicly shared models. It also features Aegis, a firewall that can detect prompt injections, leaks of prompts and PII, toxic language, and other issues.

DeepFake Detector is a tool that enables users to identify and confirm the genuineness of videos and audios. It is intended to assist users in defending themselves against deepfakes, which are AI-created videos and audios meant to disseminate false information and mislead individuals. The detector employs professional verification services to assess the likelihood of a video or audio being an AI-generated deepfake or a genuine recording. This provides users with reassurance and ensures the authenticity of audio and video content.

AICheatCheck is a learning resource aimed at assisting educators in identifying content created by AI. It achieves this by examining the structure of sentences and the text's readability to detect the presence of a GPT model effectively. At present, the tool is only compatible with the English language and works best on text exceeding 50 words.