{"podcast":{"title":"Practical AI","slug":"practical-ai","podcast_index_feed_id":444526,"rss_url":"https://feeds.transistor.fm/practical-ai-machine-learning-data-science-llm","website_url":"https://practicalai.fm","image_url":"https://img.transistorcdn.com/WMlp2ug34XB6LDJ3-vnzti_-_y144LUlFW0Xzzn3fss/rs:fill:0:0:1/w:1400/h:1400/q:60/mb:500000/aHR0cHM6Ly9pbWct/dXBsb2FkLXByb2R1/Y3Rpb24udHJhbnNp/c3Rvci5mbS8wMTZi/ZWJmNWIwNDdmYTcw/NGJjMTExZjNjZmYy/M2ZjNS5wbmc.jpg","author":"Practical AI LLC","episode_count":357,"summary":"Making artificial intelligence practical, productive & accessible to everyone. Practical AI is a show in which technology professionals, business people, students, enthusiasts, and expert guests engage in lively discussions about Artificial Intelligence and related topics (Machine Learning, Deep Learning, Neural Networks, GANs, MLOps, AIOps, LLMs & more). The focus is on productive implementations and real-world scenarios that are accessible to everyone. If you want to keep up with the latest advances in AI, while keeping one foot in the real world, then this is the show for you!","last_synced_at":null,"page_url":"https://stenobird.com/podcast/practical-ai"},"episode":{"title":"Controlling AI Models from the Inside","slug":"controlling-ai-models-from-the-inside","published_at":"2026-01-20T19:10:20+00:00","page_url":"https://stenobird.com/podcast/practical-ai/controlling-ai-models-from-the-inside","show_page_url":"https://stenobird.com/podcast/practical-ai","url":"https://share.transistor.fm/s/df33214d","audio_url":"https://pscrb.fm/rss/p/dts.podtrac.com/redirect.mp3/media.transistor.fm/df33214d/9c2dd1a8.mp3","summary":"Traditional AI safety relies on external filters that monitor prompts and responses, often creating latency and high costs. This episode explores a model-native approach using runtime instrumentation to detect problematic neuron activation inside the 'black box' before bad outputs are even generated.","meta_description":"Explore model-native AI safety: moving beyond prompt filters to runtime instrumentation and internal model visibility with Alizishaan Khatri.","key_points":["Main idea: Current AI safety is limited to the 'gatekeeper' layer, analyzing only inputs and outputs","Failure mode: External guardrails can be bypassed by jailbreaks and are often too expensive or slow for production","Practical takeaway: Monitoring internal model subspaces allows for intervention during the generation process, not just after","Technical concept: Model-native safety involves instrumenting the model to identify specific subregions that trigger during toxic or unauthorized content generation","Future vision: Creating a standardized safety layer that enables the use of LLMs in highly regulated industries like healthcare"],"chapters":[{"start_ms":60000,"title":"Introduction","summary":"Hosts Daniel and Chris introduce Alizishaan Khatri, founder of Wrynx, and set the stage for discussing the future of AI model safety."},{"start_ms":260000,"title":"AI for Security vs. Security for AI","summary":"Distinguishing between using AI to solve security problems and the challenge of securing the AI models themselves as they enter the tech stack."},{"start_ms":445000,"title":"The Limits of Prompt Filtering","summary":"An analysis of why current 'gatekeeper' solutions—analyzing prompts and responses—are insufficient against sophisticated jailbreaks."},{"start_ms":1065000,"title":"Model-Native Instrumentation","summary":"Exploring the concept of 'cameras inside the building' by monitoring internal model subspaces and neuron activation at runtime."},{"start_ms":1455000,"title":"The Burden of Custom Training","summary":"Discussing why customers cannot simply train new models to avoid certain topics and the need for a more scalable safety layer."},{"start_ms":2030000,"title":"Detecting Toxicity via Subspaces","summary":"How identifying specific model regions that trigger during toxic generation allows for proactive intervention."},{"start_ms":2435000,"title":"The Future of Model Safety","summary":"Alizishaan outlines his vision for a de facto safety layer that enables LLM adoption in sensitive sectors like healthcare."}],"topics":["AI Safety","Large Language Models","Model Interpretability","Runtime Security","AI Guardrails","Machine Learning Infrastructure","Cybersecurity","AI Governance"],"duration_seconds":2635,"processing_state":"processed","actions":[{"name":"request_transcript","method":"POST","url":"https://stenobird.com/v1/public/podcasts/practical-ai/episodes/controlling-ai-models-from-the-inside/transcription-requests","description":"Idempotently request low-priority transcript generation for this episode."},{"name":"read_markdown","method":"GET","url":"https://stenobird.com/podcast/practical-ai/controlling-ai-models-from-the-inside.md","description":"Read the agent-friendly Markdown representation of this episode resource."}]}}