This is a submission to JIT HACK 24 at JYOTHY INSTITUTE OF TECHNOLOGY

PROBLEM STATEMENT: Deepfake technology poses significant risks to media authenticity and personal security. Our challenge is to develop a platform that effectively detects and protects against deepfakes, ensuring trust in digital content through AI, user validation, and decentralized technologies.

OUR SOLUTION: We make use of AI models to protect images and audios from manipulation by AI through introduction of noise and other factors into the media thus making it unprocessable by AI Models. We also make use of AI models to detect AI modified audios and videos using CNN models. We also use blockchain to introduce a voting system so that people can vote on wheather an Image is AI generated or real. We use solidity for smart contract, polygon testnet and metamask.

TECH STACK: FRONT END: Next js BACKEND: python and flask STORAGE: IPFS AI/ML: Tensorflow, keras, librosa, torch

FLOW OF PROCESS: img