<![CDATA[Deep Moat]]>When AI rewrites the rules of distribution, pricing, and product, what still counts as a moat? Deep Moat is a podcast about AI and competitive strategy, with founders, operators, and investors on what defensibility actually looks like.

]]>https://riverside.comRiverside.fm (https://riverside.com)Sat, 18 Jul 2026 03:39:38 GMTMon, 04 May 2026 13:19:31 GMT60Lucian Armasu & John Komarek<p>When AI rewrites the rules of distribution, pricing, and product, what still counts as a moat? Deep Moat is a podcast about AI and competitive strategy, with founders, operators, and investors on what defensibility actually looks like.</p>episodicLucian Armasu & John Komareklucian@adslux.comno<![CDATA[Grok 4.5 and Musk's Impenetrable Moat]]>Grok 4.5 puts out around four times fewer tokens than GPT 5.5 or Opus 4.8, which is most of why it lands at four times the price cut. So if SpaceX AI has the GPUs, the cursor team and the rocket scientists, what is actually still holding them back?

This week on Deep Moat:

◼ Why Grok 4.5's token efficiency, not its benchmarks, is the real story

◼ Musk fires the Grok team, brings in SpaceX engineers, and buys cursor for the coding data

◼ The trillion parameter roadmap: 1.5 now, then 2, 3, 6, and 10

◼ Why catching Anthropic and OpenAI probably waits for Grok 5

◼ The uncomfortable read that SpaceX overtaking the top two knocks Google out of the top three

◼ Google's dark times, again, and why Gemini 3.5 Flash shipped while Pro slipped eight months

◼ Why a Gmail agent doesn't need the smartest model, just good enough, fast and cheap

◼ SpaceX holds nearly half the AI chips rolled out globally, and Musk is talking about a Terafab in orbit

Chapters:

0:00 Intro

0:28 Grok 4.5 and the token drop

2:00 Cost and speed as positioning

3:57 Why Grok fell behind

4:44 Musk fires the team, buys cursor

6:23 Why cursor's data matters

8:12 The trillion parameter roadmap

9:22 Does this push Google out?

10:48 Google's dark times

14:56 Google bets on AI search

16:56 SpaceX's chip moat

17:50 Chips in space

]]>3e2c6d0c-a105-40fc-861c-fd053504c65cThu, 16 Jul 2026 12:27:47 GMT<p>Grok 4.5 puts out around four times fewer tokens than GPT 5.5 or Opus 4.8, which is most of why it lands at four times the price cut. So if SpaceX AI has the GPUs, the cursor team and the rocket scientists, what is actually still holding them back?</p><p></p><p>This week on Deep Moat:</p><p>◼ Why Grok 4.5's token efficiency, not its benchmarks, is the real story</p><p>◼ Musk fires the Grok team, brings in SpaceX engineers, and buys cursor for the coding data</p><p>◼ The trillion parameter roadmap: 1.5 now, then 2, 3, 6, and 10</p><p>◼ Why catching Anthropic and OpenAI probably waits for Grok 5</p><p>◼ The uncomfortable read that SpaceX overtaking the top two knocks Google out of the top three</p><p>◼ Google's dark times, again, and why Gemini 3.5 Flash shipped while Pro slipped eight months</p><p>◼ Why a Gmail agent doesn't need the smartest model, just good enough, fast and cheap</p><p>◼ SpaceX holds nearly half the AI chips rolled out globally, and Musk is talking about a Terafab in orbit</p><p></p><p>Chapters:</p><p>0:00 Intro</p><p>0:28 Grok 4.5 and the token drop</p><p>2:00 Cost and speed as positioning</p><p>3:57 Why Grok fell behind</p><p>4:44 Musk fires the team, buys cursor</p><p>6:23 Why cursor's data matters</p><p>8:12 The trillion parameter roadmap</p><p>9:22 Does this push Google out?</p><p>10:48 Google's dark times</p><p>14:56 Google bets on AI search</p><p>16:56 SpaceX's chip moat</p><p>17:50 Chips in space</p>no00:18:29Grok 4.5 and Musk's Impenetrable Moatfull<![CDATA[Fable 5 Is Too DANGEROUS for You (So They Removed It)]]>Anthropic's new top model, Fable 5, is smarter than anything else out there, so they're locking parts of it down, pulling it out of subscriptions, and charging double for it. Is this the most powerful AI yet, or the moment AI starts splitting into haves and have-nots?

This week on Deep Moat:

◼ Why Fable 5 sits in a whole new tier above Opus, and caught OpenAI off guard

◼ The cybersecurity and biochem restrictions that have people calling it "too controlled"

◼ Fable 5's effort levels, and why low effort already beats Opus 4.8 at high

◼ Pulling Fable from subscriptions after June 22, and the high-risk bet on per-token pricing

◼ Uber and Microsoft capping how much AI each employee can spend

◼ Price per outcome vs price per token, and the Stripe rebuild that took 1 day instead of 2 months

◼ Where Fable still loses: GPT 5.5 wins the vending benchmark, cheaper and faster on coding

◼ The real worry: AI for the elite, two classes, and "getting our AI democracy from China"

Chapters:

0:00 Intro & Fable 5 release

1:56 Hands-on: one-shotting & design

3:08 A new tier above Opus

4:54 Switching to Codex & GPT 5.5

7:01 The cost problem

9:06 Price per outcome, not tokens

11:56 Picking the right model

13:06 Effort levels explained

14:02 Benchmarks: vending & coding

16:21 Is Anthropic gatekeeping AI?

]]>164955cf-c6b0-4c45-98ad-57065a59c8b3Mon, 15 Jun 2026 11:16:36 GMT<p>Anthropic's new top model, Fable 5, is smarter than anything else out there, so they're locking parts of it down, pulling it out of subscriptions, and charging double for it. Is this the most powerful AI yet, or the moment AI starts splitting into haves and have-nots?</p><p></p><p>This week on Deep Moat:</p><p>◼ Why Fable 5 sits in a whole new tier above Opus, and caught OpenAI off guard</p><p>◼ The cybersecurity and biochem restrictions that have people calling it "too controlled"</p><p>◼ Fable 5's effort levels, and why low effort already beats Opus 4.8 at high</p><p>◼ Pulling Fable from subscriptions after June 22, and the high-risk bet on per-token pricing</p><p>◼ Uber and Microsoft capping how much AI each employee can spend</p><p>◼ Price per outcome vs price per token, and the Stripe rebuild that took 1 day instead of 2 months</p><p>◼ Where Fable still loses: GPT 5.5 wins the vending benchmark, cheaper and faster on coding</p><p>◼ The real worry: AI for the elite, two classes, and "getting our AI democracy from China"</p><p></p><p>Chapters:</p><p>0:00 Intro &amp; Fable 5 release</p><p>1:56 Hands-on: one-shotting &amp; design</p><p>3:08 A new tier above Opus</p><p>4:54 Switching to Codex &amp; GPT 5.5</p><p>7:01 The cost problem</p><p>9:06 Price per outcome, not tokens</p><p>11:56 Picking the right model</p><p>13:06 Effort levels explained</p><p>14:02 Benchmarks: vending &amp; coding</p><p>16:21 Is Anthropic gatekeeping AI?</p>no00:18:17Fable 5 Is Too DANGEROUS for You (So They Removed It)full<![CDATA[Opus 4.8 Just Exposed Anthropic's Biggest Problem]]>Opus 4.8 is 61% cheaper than 4.7 and a bit smarter on most benchmarks, but it lost something on the way there. This week we break down what Anthropic traded for that price drop, why everyone in business suddenly talks about Claude instead of ChatGPT, and what Uber's $1,500/month per-employee AI cap signals about where enterprise spending is headed.

We also get into the new effort levels in Claude, whether Opus 4.x has hit a ceiling that only a new architecture can break, NVIDIA's Vera CPU built specifically for agentic workflows, and why "half your customers being AI agents" is closer than most businesses are planning for.

Chapters:

0:00 Intro & Opus 4.8 release

2:58 Why Claude lost its personality

4:19 Picking the right effort level

5:16 Has Opus 4.x hit a ceiling?

6:25 Claude is overtaking ChatGPT

8:49 Claude for work, ChatGPT for personal?

11:35 Uber's $1,500 AI spend cap

14:30 How to actually measure AI ROI

16:18 NVIDIA's Vera CPU for agents

17:21 Agent-to-agent payments are coming

]]>367fc042-9afd-4c4a-9952-0087275b8e16Fri, 05 Jun 2026 22:06:48 GMT<p>Opus 4.8 is 61% cheaper than 4.7 and a bit smarter on most benchmarks, but it lost something on the way there. This week we break down what Anthropic traded for that price drop, why everyone in business suddenly talks about Claude instead of ChatGPT, and what Uber's $1,500/month per-employee AI cap signals about where enterprise spending is headed.</p><p></p><p>We also get into the new effort levels in Claude, whether Opus 4.x has hit a ceiling that only a new architecture can break, NVIDIA's Vera CPU built specifically for agentic workflows, and why "half your customers being AI agents" is closer than most businesses are planning for.</p><p></p><p>Chapters:</p><p>0:00 Intro &amp; Opus 4.8 release</p><p>2:58 Why Claude lost its personality</p><p>4:19 Picking the right effort level</p><p>5:16 Has Opus 4.x hit a ceiling?</p><p>6:25 Claude is overtaking ChatGPT</p><p>8:49 Claude for work, ChatGPT for personal?</p><p>11:35 Uber's $1,500 AI spend cap</p><p>14:30 How to actually measure AI ROI</p><p>16:18 NVIDIA's Vera CPU for agents</p><p>17:21 Agent-to-agent payments are coming</p>no00:19:20Opus 4.8 Just Exposed Anthropic's Biggest Problemfull<![CDATA[Microsoft Just KILLED Claude Code (It's Bigger Than You Think)]]>Microsoft just killed Claude Code inside the company, and Nvidia, Meta, and Uber are all rethinking how much they let employees burn on AI coding. The "token maxing" era is ending fast.

In this episode of DeepMoat we break down what that means for how teams should actually use AI in 2026: Microsoft's pivot to GitHub Copilot, why Uber torched a full year of API budget in three months, and the real reason behind the "use 50% of your salary on AI" memos.

We also get into Opus 4.7 second impressions, where it finally clicks (design work, presentations, day-to-day vs GPT-5.5), and why Anthropic's stale training data still bites in coding work.

Then we go deep on AI agents: what they're actually useful for today, how to use the Buy Back Your Time framework to pick which tasks to automate, why evals matter before you trust an agent's output, and why low-stakes repetitive work is still the sweet spot until hallucination rates drop further.

If you're a founder, engineer, or operator trying to figure out where AI coding tools and AI agents actually drive ROI versus where they burn money, this one's for you.

Chapters

0:00 Intro

0:24 Microsoft kills Claude Code

1:27 Nvidia and Meta's token-maxing

4:08 Why Microsoft pushed Copilot

4:42 Opus 4.7 second impressions

7:43 Anthropic's stale training data

8:50 Why co-work still frustrates

10:00 Opus 4.7 vs GPT-5.5 for slides

13:30 AI agents: hype vs reality

15:35 Picking the right tasks to automate

18:56 Low-stakes is the sweet spot

#ClaudeCode #AIAgents #AICoding

]]>b19f292a-62c1-412a-ab99-dd601f496871Fri, 29 May 2026 21:49:40 GMT<p>Microsoft just killed Claude Code inside the company, and Nvidia, Meta, and Uber are all rethinking how much they let employees burn on AI coding. The "token maxing" era is ending fast.</p><p></p><p>In this episode of DeepMoat we break down what that means for how teams should actually use AI in 2026: Microsoft's pivot to GitHub Copilot, why Uber torched a full year of API budget in three months, and the real reason behind the "use 50% of your salary on AI" memos.</p><p></p><p>We also get into Opus 4.7 second impressions, where it finally clicks (design work, presentations, day-to-day vs GPT-5.5), and why Anthropic's stale training data still bites in coding work.</p><p></p><p>Then we go deep on AI agents: what they're actually useful for today, how to use the Buy Back Your Time framework to pick which tasks to automate, why evals matter before you trust an agent's output, and why low-stakes repetitive work is still the sweet spot until hallucination rates drop further.</p><p></p><p>If you're a founder, engineer, or operator trying to figure out where AI coding tools and AI agents actually drive ROI versus where they burn money, this one's for you.</p><p></p><p>Chapters</p><p>0:00 Intro</p><p>0:24 Microsoft kills Claude Code</p><p>1:27 Nvidia and Meta's token-maxing</p><p>4:08 Why Microsoft pushed Copilot</p><p>4:42 Opus 4.7 second impressions</p><p>7:43 Anthropic's stale training data</p><p>8:50 Why co-work still frustrates</p><p>10:00 Opus 4.7 vs GPT-5.5 for slides</p><p>13:30 AI agents: hype vs reality</p><p>15:35 Picking the right tasks to automate</p><p>18:56 Low-stakes is the sweet spot</p><p></p><p>#ClaudeCode #AIAgents #AICoding</p>no00:21:15Microsoft Just KILLED Claude Code (It's Bigger Than You Think)full<![CDATA[Google I/O Killed SEO]]>Google just shipped enough at I/O to become the Apple of AI. We get into Gemini 3.5 Flash beating last year's Pro, why Spark is built so you never leave Google, and how Universal Cart quietly killed SEO.

In this episode:

  • What Gemini 3.5 Flash actually changes (and why it costs more)
  • Why Spark is the lock-in play, not the agent play
  • Gemini Omni vs Seedance and why everyone misses the point
  • Universal Cart and what's left of search traffic
  • Why Karpathy picked Anthropic over everyone

Timestamps

00:00 Intro

00:26 Gemini 3.5 Flash beats the old Pro

04:22 Why every AI model is getting more expensive

06:41 Gemini Spark and the OpenClaw question

12:14 Gemini Omni vs Seedance

16:08 Universal Cart and the death of SEO

22:34 Karpathy joins Anthropic

]]>257a4d40-4eb9-4dfa-8b8b-dd9894addc65Fri, 22 May 2026 21:19:08 GMT<p>Google just shipped enough at I/O to become the Apple of AI. We get into Gemini 3.5 Flash beating last year's Pro, why Spark is built so you never leave Google, and how Universal Cart quietly killed SEO.</p><p></p><p><b>In this episode:</b></p><ul><li>What Gemini 3.5 Flash actually changes (and why it costs more)</li><li>Why Spark is the lock-in play, not the agent play</li><li>Gemini Omni vs Seedance and why everyone misses the point</li><li>Universal Cart and what's left of search traffic</li><li>Why Karpathy picked Anthropic over everyone</li></ul><p></p><p><b>Timestamps </b></p><p>00:00 Intro</p><p>00:26 Gemini 3.5 Flash beats the old Pro</p><p>04:22 Why every AI model is getting more expensive</p><p>06:41 Gemini Spark and the OpenClaw question</p><p>12:14 Gemini Omni vs Seedance</p><p>16:08 Universal Cart and the death of SEO</p><p>22:34 Karpathy joins Anthropic</p>no00:24:56Google I/O Killed SEOfull<![CDATA[AI's Small Business Problem (and Shopify's Bigger One)]]>AI is quietly eating ecommerce, and Shopify's running out of moat. We get into Anthropic's new Cloud for Business, why 96% of AI deployments still fail, and how Stripe just made it possible for an agent to spin up a whole stack from a single command.

In this episode:

  • What Cloud for Business actually is (and whether it solves anything)
  • Why forward deployed engineers became tech's hottest role
  • The case for ditching Shopify and wipe-coding your storefront with Claude Code
  • Stripe Projects and the rise of agent-native infrastructure
  • Why MCP servers are about to be unavoidable

Timestamps

00:00 Claude for Business: what Anthropic just launched

04:25 Forward deployed engineers and the 96% failure rate

09:01 Is Shopify in real trouble?

12:46 What's left of Shopify's moat

15:41 Stripe Projects and agent-native infrastructure

19:25 Why MCP servers are about to be table stakes

]]>3fd28dce-4ddf-4886-abd1-806c9552473eTue, 19 May 2026 14:42:05 GMT<p>AI is quietly eating ecommerce, and Shopify's running out of moat. We get into Anthropic's new Cloud for Business, why 96% of AI deployments still fail, and how Stripe just made it possible for an agent to spin up a whole stack from a single command.</p><p></p><p>In this episode:</p><ul><li>What Cloud for Business actually is (and whether it solves anything)</li><li>Why forward deployed engineers became tech's hottest role</li><li>The case for ditching Shopify and wipe-coding your storefront with Claude Code</li><li>Stripe Projects and the rise of agent-native infrastructure</li><li>Why MCP servers are about to be unavoidable</li></ul><p></p><p><b>Timestamps</b></p><p>00:00 Claude for Business: what Anthropic just launched</p><p>04:25 Forward deployed engineers and the 96% failure rate</p><p>09:01 Is Shopify in real trouble?</p><p>12:46 What's left of Shopify's moat</p><p>15:41 Stripe Projects and agent-native infrastructure</p><p>19:25 Why MCP servers are about to be table stakes</p>no00:20:08AI's Small Business Problem (and Shopify's Bigger One)full<![CDATA[Anthropic Pulls Ahead: Claude Code, Codex, and the AI Agent Wars]]>Anthropic is pulling ahead, OpenAI is fighting back, and the AI agent wars are getting messy.

In this episode, we break down Anthropic’s enterprise-first strategy, why Claude Code became a serious growth engine, and how OpenAI’s Codex comeback could shift the race again.

We also dig into Anthropic’s surprising SpaceX compute deal, the growing pains behind OpenClaw, and why the future of AI work may not be one giant agent - but a smarter orchestration layer of specialized agents.

Takeaways

  • Anthropic’s enterprise-first strategy is paying off
    Claude Code vs Codex is becoming the key developer battleground
  • The AI market is moving toward “service as software”
    Anthropic’s SpaceX compute deal could ease Claude’s capacity issues
  • OpenClaw’s instability may be pushing users toward Claude Code and Codex
  • The future may be unified interfaces with specialized sub-agents

Chapters

00:00 Intro

00:22 Anthropic Pulls Ahead of OpenAI
07:05 Anthropic’s SpaceX Compute Deal
10:04 OpenClaw vs Claude Code
16:03 The Future of AI Agent UX

]]>4e6dcf37-e9a8-4e27-a657-90c852fdb4b6Tue, 12 May 2026 18:52:47 GMT<p>Anthropic is pulling ahead, OpenAI is fighting back, and the AI agent wars are getting messy.</p><p></p><p>In this episode, we break down Anthropic’s enterprise-first strategy, why Claude Code became a serious growth engine, and how OpenAI’s Codex comeback could shift the race again. </p><p></p><p>We also dig into Anthropic’s surprising SpaceX compute deal, the growing pains behind OpenClaw, and why the future of AI work may not be one giant agent - but a smarter orchestration layer of specialized agents.</p><p></p><p><b>Takeaways</b></p><ul><li>Anthropic’s enterprise-first strategy is paying off<br />Claude Code vs Codex is becoming the key developer battleground</li><li>The AI market is moving toward “service as software”<br />Anthropic’s SpaceX compute deal could ease Claude’s capacity issues</li><li>OpenClaw’s instability may be pushing users toward Claude Code and Codex</li><li>The future may be unified interfaces with specialized sub-agents</li></ul><p></p><p><b>Chapters</b></p><p>00:00 Intro</p><p>00:22 Anthropic Pulls Ahead of OpenAI<br />07:05 Anthropic’s SpaceX Compute Deal<br />10:04 OpenClaw vs Claude Code<br />16:03 The Future of AI Agent UX</p>no00:28:49Anthropic Pulls Ahead: Claude Code, Codex, and the AI Agent Warsfull<![CDATA[GPT-5.5 Gets Sharper, Meta Opens the Ads Stack]]>The conversation covers the improvements in GPT 5.5, the shift in model autonomy and specificity, and Meta's acquisition of Manus and the release of CLI and MCP server access for Facebook AI management.

Takeaways

  • GPT 5.5 improvements
  • Model autonomy and specificity
  • Meta's acquisition of Manus
  • Meta Ads CLI and MCP Server access

Chapters

  • 00:00 GPT 5.5 Improvements
  • 08:54 Meta's Acquisition of Manus
  • 12:12 CLI and MCP Server Access
]]>fdb36044-2fff-4e63-adcc-49e20acf9af3Mon, 04 May 2026 13:27:02 GMT<p>The conversation covers the improvements in GPT 5.5, the shift in model autonomy and specificity, and Meta's acquisition of Manus and the release of CLI and MCP server access for Facebook AI management.</p><p></p><p>Takeaways</p><ul><li>GPT 5.5 improvements</li><li>Model autonomy and specificity</li><li>Meta's acquisition of Manus</li><li>Meta Ads CLI and MCP Server access</li></ul><p></p><p>Chapters</p><ul><li>00:00 GPT 5.5 Improvements</li><li>08:54 Meta's Acquisition of Manus</li><li>12:12 CLI and MCP Server Access</li></ul>no00:15:05GPT-5.5 Gets Sharper, Meta Opens the Ads Stackfull