In this episode I cover how AI access and permissions have changed since ChatGPT launched back in 2022, from zero connections and basically zero worry to the current YOLO approach of vibe coders and early adopters. I share what I currently give Claude access to, what my biggest concern is, and offer a read/write/act framework that you can use to decide what your AI should touch.

• How AI access sentiment has changed over 4 years
• 2022: ChatGPT with zero connections
• 2023: Bard connects to Gmail, Docs, Drive, Maps
• MCP and the "AI in my email" era
• Agentic AI, Claude Code, and Claude Cowork
• What I currently give Claude access to
• Prompt injection: my biggest concern
• Read/write/act framework
• Privacy, client data, and benefit vs. risk
• How I used AI this week: Claude-Descript connector and pulling quotes from transcripts

• Listen to Ep. 51: WTF is MCP
• Read this episode's Curious Companion
• Subscribe to the Prompting Curiosity newsletter
• Submit a Question
• Visit the Website
• Feeling curious AND generous? Click here to support the podcast.

• (00:00:05) - Intro and Episode Overview
• (00:01:32) - Early AI Access: ChatGPT and Bard Origins (2022-2023)
• (00:04:16) - MCP and the Rise of Agentic AI (2024-2025)
• (00:06:23) - Claude Cowork and Mainstream Agentic Adoption
• (00:08:23) - Personal Evolution in Granting AI Access
• (00:10:15) - Prompt Injection Risks Explained
• (00:11:20) - A Framework: Read, Write, Act and Whose Data
• (00:13:38) - Privacy Concerns vs Personal Comfort
• (00:15:01) - Training Data Concerns
• (00:15:28) - Handling Clients' and Others' Data
• (00:17:08) - Healthcare, Therapy, and Sensitive Data Gray Areas
• (00:18:34) - Benefit vs Risk: The Big Picture on AI Access
• (00:20:26) - Final Thoughts on What Your AI Should Access
• (00:20:58) - How I Used AI This Week: Claude and Descript
• (00:23:10) - Outro and Sign Off

Is AI actually going to kill us all?There's been a ton of talk about this topic in the media lately, and so I'm dedicating an episode to sharing my thoughts. Spoiler alert, no, I don’t think that AI is going to kill us, but that doesn’t mean that this technology has no risk. I walk through the Jacob Coxon resignation thread and the fact that Anthropic's own alignment team admits there's genuine risk, then get into why the doom talk feels more irresponsible than credible. I don't think regulation is coming to save us anytime soon, but to me, the answer isn't fear, it’s curiosity.

• Directly answering the question: Will AI kill us all?
• Jacob Coxon's resignation thread on X
• Evan Hubinger's tweets on AI risk
• Defining superintelligence and recursive self-improvement (RSI)
• Moving the goalposts from AGI to superintelligence
• AI's coding and hacking capabilities
• Jake Handy's prediction about a coming AI crisis
• Why the AI-doom talk feels irresponsible
• Regulation and data centers as oil
• Dario Amodei's blog post
• Change doesn't happen in a vacuum
• How I used AI this week: Replacing my Jeep’s thermostat housing

• Read Jacob Coxon's resignation thread on X
• Listen to Ed Zitron's podcast episode about AI and regulation
• Read Dario Amodei's blog post, "We Must Pace the Frontier"
• Read this episode's Curious Companion
• Subscribe to the Prompting Curiosity newsletter
• Submit a Question
• Visit the Website
• Feeling curious AND generous? Click here to support the podcast.

• (00:00:05) - Intro and today's topic: will AI kill us all?
• (00:03:29) - Background: Jacob Cox's resignation and Anthropic's Evan Hubinger
• (00:05:18) - Defining superintelligence and recursive self-improvement
• (00:07:19) - Why the doom claims seem overblown right now
• (00:10:39) - Hacking risks and predictions of a coming AI-caused crisis
• (00:12:15) - Irresponsible messaging and the case for transparency
• (00:13:24) - Government incentives, profit motives, and the China narrative
• (00:17:24) - Dario Amodei's regulation plan and skepticism about government action
• (00:19:10) - Staying a realist: humans, open source, and hope for the future
• (00:21:11) - How I Use AI This Week: the rattling thermostat story
• (00:23:17) - Closing thoughts and sign-off

In this episode I talk about the eleventy billion AI models that dropped in the past few weeks, including Anthropic's Fable 5.1 and Mythos 5.1, OpenAI's GPT-6 Astra, and Meta's new Muse agent. I get into former Anthropic researcher Jacob Coxon's resignation and his warning about AI risk, plus what it actually means that OpenAI just rated Astra "Critical" for cybersecurity capability. I also share why I think Meta Muse is a terrible idea and what all of these model drops say about the direction AI is heading.

• Jacob Coxon's resignation and AI regulation warnings
• Anthropic's Fable 5.1 and Mythos 5.1 updates
• OpenAI's GPT-6 Astra release
• Astra's "Critical" cybersecurity capability rating
• AGI throwback to Episode 7
• Meta Muse launch: everyone gets an agent
• Meta Muse access, pricing, and privacy details
• Why Meta Muse is a terrible idea
• How I used AI this week: Gemini AI Overviews

• Read Jacob's statements on X
• Listen to Ep. 7: What is AI?
• Subscribe to the Prompting Curiosity newsletter
• Submit a Question
• Visit the Website
• Feeling curious AND generous? Click here to support the podcast.

• (00:00:05) - Intro and welcome
• (00:00:38) - Jacob Coxon's anthropic resignation statement
• (00:01:41) - Reflections on AI regulation and motives
• (00:04:56) - Recent model drops overview
• (00:06:14) - Anthropic's Fable and Mythos updates
• (00:07:25) - OpenAI's GPT-6 Astra and computer use capabilities
• (00:10:01) - Astra's critical cyber capability rating
• (00:11:27) - AGI claims and the race to super intelligence
• (00:13:46) - Meta's Muse agent launch and concerns
• (00:15:13) - Muse's privacy and data risks
• (00:16:44) - Who Meta Muse targets and risks to users
• (00:17:38) - Companies' true priorities: money over safety
• (00:18:23) - How I used AI this week: Gemini overviews
• (00:20:56) - Wrap up and outro