{"type":"video","version":"1.0","html":"<iframe src=\"https://www.loom.com/embed/ced7c9415bd945fd9cce264c45e08d01\" frameborder=\"0\" width=\"1662\" height=\"1246\" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>","height":1246,"width":1662,"provider_name":"Loom","provider_url":"https://www.loom.com","thumbnail_height":1246,"thumbnail_width":1662,"thumbnail_url":"https://cdn.loom.com/sessions/thumbnails/ced7c9415bd945fd9cce264c45e08d01-73e450f5fd9d2077.gif","duration":260.026,"title":"Introducing Talos: Decentralized Adversarial AI Security on Bittensor","description":"Hello everyone, in this video, I introduced Talos, our decentralized adversarial AI security subnet on Bitensor, where red miners generate attack prompts and blue miners defend against them. We create a competitive environment that rewards both attackers and defenders, with validators orchestrating the scoring process based on attack severity and classification accuracy. I also walked you through our testnet implementation, which currently features one validator, one blue miner, and one red miner due to API limitations. I encourage you to try launching the campaign on our local host and check the Threat Intelligence Dashboard for updates on prompt statuses. Thank you for your attention!"}