Search a title or topic

Over 20 million podcasts, powered by 

Player FM logo
Artwork

Content provided by Gigi and Pablo. All podcast content including episodes, graphics, and podcast descriptions are uploaded and provided directly by Gigi and Pablo or their podcast platform partner. If you believe someone is using your copyrighted work without your permission, you can follow the process outlined here https://podcastplayer.com/legal.
Player FM - Podcast App
Go offline with the Player FM app!

#10 - Walking with Jesus

1:36:12
 
Share
 

Manage episode 478527790 series 3652926
Content provided by Gigi and Pablo. All podcast content including episodes, graphics, and podcast descriptions are uploaded and provided directly by Gigi and Pablo or their podcast platform partner. If you believe someone is using your copyrighted work without your permission, you can follow the process outlined here https://podcastplayer.com/legal.

"It's gonna be permissionless or hell."

Gigi and gzuuus are vibing towards dystopia.

Books & articles mentioned:

In this dialogue:

  • Pablo's Roo Setup
  • Tech Hype Cycles
  • AI 2027
  • Prompt injection and other attacks
  • Goose and DVMCP
  • Cursor vs Roo Code
  • Staying in control thanks to Amber and signing delegation
  • Is YOLO mode here to stay?
  • What agents to trust?
  • What MCP tools to trust?
  • What code snippets to trust?
  • Everyone will run into the issues of trust and micropayments
  • Nostr solves Web of Trust & micropayments natively
  • Minimalistic & open usually wins
  • DVMCP exists thanks to Totem
  • Relays as Tamagochis
  • Agents aren't nostr experts, at least not right now
  • Fix a mistake once & it's fixed forever
  • Giving long-term memory to LLMs
  • RAG Databases signed by domain experts
  • Human-agent hybrids & Chess
  • Nostr beating heart
  • Pluggable context & experts
  • "You never need an API key for anything"
  • Sats and social signaling
  • Difficulty-adjusted PoW as a rare-limiting mechanism
  • Certificate authorities and centralization
  • No solutions to policing speech!
  • OAuth and how it centralized
  • Login with nostr
  • Closed vs open-source models
  • Tiny models vs large models
  • The minions protocol (Stanford paper)
  • Generalist models vs specialized models
  • Local compute & encrypted queries
  • Blinded compute
  • "In the eyes of the state, agents aren't people"
  • Agents need identity and money; nostr provides both
  • "It's gonna be permissionless or hell"
  • We already have marketplaces for MCP stuff, code snippets, and other things
  • Most great stuff came from marketplaces (browsers, games, etc)
  • Zapstore shows that this is already working
  • At scale, central control never works. There's plenty scams and viruses in the app stores.
  • Using nostr to archive your user-generated content
  • HAVEN, blossom, novia
  • The switcharoo from advertisements to training data
  • What is Truth?
  • What is Real?
  • "We're vibing into dystopia"
  • Who should be the arbiter of Truth?
  • First Amendment & why the Logos is sacred
  • Silicon Valley AI bros arrogantly dismiss wisdom and philosophy
  • Suicide rates & the meaning crisis
  • Are LLMs symbiotic or parasitic?
  • The Amish got it right
  • Are we gonna make it?
  • Careless People by Sarah Wynn-Williams
  • Takedown by Laila michelwait
  • Harry Potter dementors & Momo's time thieves
  • Facebook & Google as non-human (superhuman) agents
  • Zapping as a conscious action
  • Privacy and the internet
  • Plausible deniability thanks to generative models
  • Google glasses, glassholes, and Meta's Ray Ben's
  • People crave realness
  • Bitcoin is the realest money we ever had
  • Nostr allows for real and honest expression
  • How do we find out what's real?
  • Constraints, policing, and chilling effects
  • Jesus' plans for DVMCP
  • Hzrd's article on how DVMs are broken (DVMs were a mistake)
  • Don't believe the hype
  • DVMs pre-date MCP tools
  • Data Vending Machines were supposed to be stupid: put coin in, get stuff out.
  • Self-healing vibe-coding
  • IP addresses as scarce assets
  • Atomic swaps and the ASS protocol
  • More marketplaces, less silos
  • The intensity of #SovEng and the last 6 weeks
  • If you can vibe-code everything, why build anything?
  • Time, the ultimate resource
  • What are the LLMs allowed to think?
  • Natural language interfaces are inherently dialogical
  • Sovereign Engineering is dialogical too

  continue reading

11 episodes

Artwork
iconShare
 
Manage episode 478527790 series 3652926
Content provided by Gigi and Pablo. All podcast content including episodes, graphics, and podcast descriptions are uploaded and provided directly by Gigi and Pablo or their podcast platform partner. If you believe someone is using your copyrighted work without your permission, you can follow the process outlined here https://podcastplayer.com/legal.

"It's gonna be permissionless or hell."

Gigi and gzuuus are vibing towards dystopia.

Books & articles mentioned:

In this dialogue:

  • Pablo's Roo Setup
  • Tech Hype Cycles
  • AI 2027
  • Prompt injection and other attacks
  • Goose and DVMCP
  • Cursor vs Roo Code
  • Staying in control thanks to Amber and signing delegation
  • Is YOLO mode here to stay?
  • What agents to trust?
  • What MCP tools to trust?
  • What code snippets to trust?
  • Everyone will run into the issues of trust and micropayments
  • Nostr solves Web of Trust & micropayments natively
  • Minimalistic & open usually wins
  • DVMCP exists thanks to Totem
  • Relays as Tamagochis
  • Agents aren't nostr experts, at least not right now
  • Fix a mistake once & it's fixed forever
  • Giving long-term memory to LLMs
  • RAG Databases signed by domain experts
  • Human-agent hybrids & Chess
  • Nostr beating heart
  • Pluggable context & experts
  • "You never need an API key for anything"
  • Sats and social signaling
  • Difficulty-adjusted PoW as a rare-limiting mechanism
  • Certificate authorities and centralization
  • No solutions to policing speech!
  • OAuth and how it centralized
  • Login with nostr
  • Closed vs open-source models
  • Tiny models vs large models
  • The minions protocol (Stanford paper)
  • Generalist models vs specialized models
  • Local compute & encrypted queries
  • Blinded compute
  • "In the eyes of the state, agents aren't people"
  • Agents need identity and money; nostr provides both
  • "It's gonna be permissionless or hell"
  • We already have marketplaces for MCP stuff, code snippets, and other things
  • Most great stuff came from marketplaces (browsers, games, etc)
  • Zapstore shows that this is already working
  • At scale, central control never works. There's plenty scams and viruses in the app stores.
  • Using nostr to archive your user-generated content
  • HAVEN, blossom, novia
  • The switcharoo from advertisements to training data
  • What is Truth?
  • What is Real?
  • "We're vibing into dystopia"
  • Who should be the arbiter of Truth?
  • First Amendment & why the Logos is sacred
  • Silicon Valley AI bros arrogantly dismiss wisdom and philosophy
  • Suicide rates & the meaning crisis
  • Are LLMs symbiotic or parasitic?
  • The Amish got it right
  • Are we gonna make it?
  • Careless People by Sarah Wynn-Williams
  • Takedown by Laila michelwait
  • Harry Potter dementors & Momo's time thieves
  • Facebook & Google as non-human (superhuman) agents
  • Zapping as a conscious action
  • Privacy and the internet
  • Plausible deniability thanks to generative models
  • Google glasses, glassholes, and Meta's Ray Ben's
  • People crave realness
  • Bitcoin is the realest money we ever had
  • Nostr allows for real and honest expression
  • How do we find out what's real?
  • Constraints, policing, and chilling effects
  • Jesus' plans for DVMCP
  • Hzrd's article on how DVMs are broken (DVMs were a mistake)
  • Don't believe the hype
  • DVMs pre-date MCP tools
  • Data Vending Machines were supposed to be stupid: put coin in, get stuff out.
  • Self-healing vibe-coding
  • IP addresses as scarce assets
  • Atomic swaps and the ASS protocol
  • More marketplaces, less silos
  • The intensity of #SovEng and the last 6 weeks
  • If you can vibe-code everything, why build anything?
  • Time, the ultimate resource
  • What are the LLMs allowed to think?
  • Natural language interfaces are inherently dialogical
  • Sovereign Engineering is dialogical too

  continue reading

11 episodes

All episodes

×
 
Loading …

Welcome to Player FM!

Player FM is scanning the web for high-quality podcasts for you to enjoy right now. It's the best podcast app and works on Android, iPhone, and the web. Signup to sync subscriptions across devices.

 

Listen to this show while you explore
Play