NFT marketplace OpenSea are updating and improving their account verification process.
Since they first appeared NFTs have experienced quite a bit of backlash for reasons that include environmental impact and fraud. Not long after they fell victim to a phishing attack, OpenSea have recognised the need to improve security and in an announcement spoke about the evolving NFT ecosystem and how imitation and plagiarism are growing issues.
NFT marketplace OpenSea have made changes in two key areas:
- An updated account verification and collection badging system that also broadens the number of creators eligible for verification.
- An automated system to identify, remove, and prevent instances of “copymints” (copies of authentic NFT content).
Account Verification and Badging
Account verification and collection badging on OpenSea are intended to help both creators and collectors by surfacing authentic accounts and content more prominently.
OpenSea are rolling out four changes to improve the system:
- An invite-based account verification system, with broader eligibility criteria
- An updated collection badge
- A new streamlined, in-product notification and application experience
- A dedicated customer support team for verification and badging, with response within 7 days
OpenSea are committed to threading the needle between removing copymints and giving space for those substantively additive remixes to prosper. So they have announced a two-part system to help with this goal:
- Image Recognition Technology – The new copymint prevention system leverages computer-vision tech to scan all NFTs on OpenSea (including new mints). The system then matches these scans against a set of authentic collections, starting with some of the most copy-minted collections — They’ll look for flips, rotations & other permutations. This set will be expanded over the coming months and detection improvements will be made regularly.
- Dedicated Human Review – Making this image recognition technology work requires dedicated “humans in the loop” who can review removal recommendations and train OpenSea’s models continuously.