{"id":"mBexVsugWIY","title":"Illia Polosukhin: Why AI Agents Need Verifiability | Token2049","description":"NEAR co-founder Illia Polosukhin joined Avi Felman on 1000x at Token2049 Singapore.\n\nIf your AI agent points at an API endpoint, “you have no idea what’s running behind it”.\n\nIt could get a smaller or quantised model, fake output, or a bash command that steals your private keys.\n\n”You want to know exactly what’s running there.”\n\n#NEARProtocol #TOKEN2049 #AIAgents #NEARAI #Crypto\n\nNEAR is the Blockchain for AI\nWebsite: https://near.org/\nNEAR AI: https://near.ai/\nIntents: https://near.org/intents\nChain Abstraction: https://www.proximity.dev/\nSharding: https://nearone.org/\n\nFor the devs\nGitHub: https://github.com/near \nDev Docs: https://docs.near.org/\nDevhub: https://devhub.near.page/ @NEARDevHub\n\nConnect with socials\nTwitter:   / nearprotocol  \nBlog: https://near.org/blog/\nMedium:   / nearprotocol  \n#NEAR #nearprotocol #NEARAI","channel_name":"NEAR Protocol","published_at":"2026-11-07T00:00:00.000Z","thumbnail_url":"https://img.youtube.com/vi/mBexVsugWIY/hqdefault.jpg","transcript_text":"No guarantees they're going to respond to you with a real model response. They can do like smaller model, okay, quantized model or they can just insert fake output completely. For example, they can run arbitrary Bosch command and this Bosch command infects your machine with a virus and steals all your private keys for example, right? That would be great, but also like you would assume that OpenAI would have safeguards against that. But it's not on OpenAI side. This is on your agent side. Okay. So you if you're running let's say Hermes and you point it at some API endpoint that API endpoint you have no idea what's running behind it. Right? I see where you're going. So so this is where verifiability of block like of blockchain technology is very useful. You want to know exactly what's running","transcript_synced_at":"2026-11-08T03:45:47.463Z","segments":[{"start_seconds":2,"text":"No guarantees they're going to respond to you with a real model response. They can do like smaller model, okay, quantized model or they can just insert","url":"https://www.youtube.com/watch?v=mBexVsugWIY&t=2s"},{"start_seconds":11,"text":"fake output completely. For example, they can run arbitrary Bosch command and this Bosch command infects your machine","url":"https://www.youtube.com/watch?v=mBexVsugWIY&t=11s"},{"start_seconds":18,"text":"with a virus and steals all your private keys for example, right? That would be great, but also like you would assume that OpenAI would have safeguards against that.","url":"https://www.youtube.com/watch?v=mBexVsugWIY&t=18s"},{"start_seconds":26,"text":"But it's not on OpenAI side. This is on your agent side. Okay.","url":"https://www.youtube.com/watch?v=mBexVsugWIY&t=26s"},{"start_seconds":30,"text":"So you if you're running let's say Hermes and you point it at some API endpoint that API endpoint you have no idea what's running behind it. Right?","url":"https://www.youtube.com/watch?v=mBexVsugWIY&t=30s"},{"start_seconds":38,"text":"I see where you're going.","url":"https://www.youtube.com/watch?v=mBexVsugWIY&t=38s"},{"start_seconds":39,"text":"So so this is where verifiability of block like of blockchain technology is very useful. You want to know exactly what's running","url":"https://www.youtube.com/watch?v=mBexVsugWIY&t=39s"}],"youtube_url":"https://youtube.com/watch?v=mBexVsugWIY","url":"https://near-pedia.xyz/videos/mBexVsugWIY"}