# Episode 84 - The AI Safety Episode

Page: https://stenobird.com/podcast/fundraisingai-6815599/episode-84-the-ai-safety-episode
Text version: https://stenobird.com/podcast/fundraisingai-6815599/episode-84-the-ai-safety-episode.md
Podcast: [FundraisingAI](https://stenobird.com/podcast/fundraisingai-6815599)
Published: 2026-04-22T10:00:00+00:00
Episode link: https://traffic.megaphone.fm/ICAAL6698791729.mp3?updated=1776841600
Audio file: https://traffic.megaphone.fm/ICAAL6698791729.mp3?updated=1776841600
Processing state: not_requested
JSON: https://stenobird.com/v1/public/podcasts/fundraisingai-6815599/episodes/episode-84-the-ai-safety-episode
Duration seconds: 2184

## Resource

What if the tools we built today decide how society advances tomorrow? With the emergence of world models, which are capable of understanding causality, basic physics, spatial relationships, and time, they are prepared to predict consequences at an extraordinary scale. Powerful tools like these have the ability to bring both enormous potential for breakthroughs in medicine, security, and innovation and amplify risks from misuse, cyberattacks, or unintended societal impacts. The responsibility rests on the shoulders of those who witness these developments at present to establish guardrails, foster collaboration, and use them with transparency and accountability. In this episode of the Fundraising AI podcast, Nathan and Scott bring awareness to AI safety, where they share opinions on setting boundaries, staying informed, and using AI transparently to align with personal values. Join with them and understand how AI is capable of creating both incredible opportunities and serious risks if not managed with caution. HIGHLIGHTS [02:05] The Rise of the World Model - The beginning of a phase where AI can predict and simulate reality itself. [07:21] The Biggest Signal - Paying attention to companies who accelerate AI releases and hold back its capabilities. [12:47] Tool Dependency - The risk of building a strategy around a single platform. [18:25] Adaptive Security - Preparing the systems and organizations when things go wrong with AI. [23:33] Risk & Responsibilities - Deciding how to use AI, where to draw the line, and who to build with. [30:02] ChatGPT Voice Mode - Nathan encourages listeners to use ChatGPT's voice mode to have long, detailed conversations about complex topics. [33:40] Prompt Hack - Scott shares a prompt hack that will help listeners to be transparent and stru…

## Actions

- request_transcript: `POST https://stenobird.com/v1/public/podcasts/fundraisingai-6815599/episodes/episode-84-the-ai-safety-episode/transcription-requests` — Idempotently request low-priority transcript generation for this episode.
- read_markdown: `GET https://stenobird.com/podcast/fundraisingai-6815599/episode-84-the-ai-safety-episode.md` — Read the agent-friendly Markdown representation of this episode resource.

A page view does not enqueue transcription. Agents should invoke `request_transcript` explicitly when they need this episode processed.

## Transcript

Full transcripts are not published on public pages unless there is a clear rights basis.