{"podcast":{"title":"\"The Cognitive Revolution\" | AI Builders, Researchers, and Live Player Analysis","slug":"the-cognitive-revolution","podcast_index_feed_id":6011783,"rss_url":"https://feeds.megaphone.fm/RINTP3108857801","website_url":"https://www.cognitiverevolution.ai/","image_url":"https://megaphone.imgix.net/podcasts/30f818da-c930-11ed-9b4b-1352ca96fb17/image/888e2c534b7c2534213c97e025646932.png?ixlib=rails-4.3.1&max-w=3000&max-h=3000&fit=crop&auto=format,compress","author":"Turpentine","episode_count":346,"summary":"A biweekly podcast where hosts Nathan Labenz and Erik Torenberg interview the builders on the edge of AI and explore the dramatic shift it will unlock in the coming years. The Cognitive Revolution is part of the Turpentine podcast network. To learn more: turpentine.co","last_synced_at":null,"page_url":"https://stenobird.com/podcast/the-cognitive-revolution"},"episode":{"title":"Bioinfohazards: Jassi Pannu on Controlling Dangerous Data from which AI Models Learn","slug":"bioinfohazards-jassi-pannu-on-controlling-dangerous-data-from-which-ai-models-learn","published_at":"2026-03-11T20:39:43+00:00","page_url":"https://stenobird.com/podcast/the-cognitive-revolution/bioinfohazards-jassi-pannu-on-controlling-dangerous-data-from-which-ai-models-learn","show_page_url":"https://stenobird.com/podcast/the-cognitive-revolution","url":"https://www.cognitiverevolution.ai/bioinfohazards-jassi-pannu-on-controlling-dangerous-data-from-which-ai-models-learn/","audio_url":"https://pdst.fm/e/mgln.ai/e/1113/pscrb.fm/rss/p/traffic.megaphone.fm/RINTP4770832217.mp3?updated=1773260206","summary":"AI models are rapidly gaining the ability to bridge the gap between theoretical biology and practical pathogen engineering. Jassi Pannu proposes a Biosecurity Data Level framework to restrict dangerous functional biological data without stifling open science.","meta_description":"Explore the intersection of AI and biosecurity with Jassi Pannu. Learn about the Biosecurity Data Level framework and strategies to prevent engineered pan…","key_points":["Main idea: Frontier AI models can provide 'uplift' by translating complex biological knowledge into actionable instructions for non-experts","Failure mode: The availability of specific mutation protocols in open literature allows for the creation of highly transmissible viruses with minimal effort","Practical takeaway: Implementing a 'Biosecurity Data Level' framework can selectively restrict dangerous sequences while preserving the benefits of open-access research","Main idea: A defense-in-depth strategy—Delay, Deter, Detect, Defend—is required to counter the speed of AI-driven biological discovery","Risk factor: The rapid growth of petabytes of unannotated sequence data creates a massive, unmonitored surface area for potential misuse"],"chapters":[{"start_ms":60000,"title":"The Risk of Gain-of-Function Research","summary":"An examination of how published research on avian influenza mutations demonstrates the ease of increasing pathogen transmissibility."},{"start_ms":1035000,"title":"Threat Actors and Data Security","summary":"Analyzing how AI model developers can implement security levels corresponding to the difficulty of preventing misuse."},{"start_ms":1520000,"title":"The Challenge of Massive Sequence Data","summary":"Discussing the risks associated with the vast, unannotated landscape of biological sequence data currently available."},{"start_ms":3000000,"title":"AI Models in Biology","summary":"A breakdown of how LLMs, bio-design tools, and foundation models like Evo act as next-token predictors for DNA and proteins."},{"start_ms":3970000,"title":"Dangerous AI Capabilities","summary":"Evaluating the potential for AI to provide dangerous 'uplift' and the ability of agents to bypass digital safeguards."},{"start_ms":4440000,"title":"Biosecurity Data Level Framework","summary":"Proposing a structured approach to controlling access to specific, high-risk biological datasets."},{"start_ms":5895000,"title":"A Vision for Global Defense","summary":"The necessity of global pathogen surveillance and 'bio-radar' systems to detect emerging threats in real-time."}],"topics":["Biosecurity","Artificial Intelligence","Synthetic Biology","Pathogen Surveillance","Genomic Data","AI Safety","Bioinformatics","Protein Design"],"duration_seconds":6192,"processing_state":"processed","actions":[{"name":"request_transcript","method":"POST","url":"https://stenobird.com/v1/public/podcasts/the-cognitive-revolution/episodes/bioinfohazards-jassi-pannu-on-controlling-dangerous-data-from-which-ai-models-learn/transcription-requests","description":"Idempotently request low-priority transcript generation for this episode."},{"name":"read_markdown","method":"GET","url":"https://stenobird.com/podcast/the-cognitive-revolution/bioinfohazards-jassi-pannu-on-controlling-dangerous-data-from-which-ai-models-learn.md","description":"Read the agent-friendly Markdown representation of this episode resource."}]}}