Ex-Amazon AI Leader: In 1 Year, the Gap Between AI Users and Everyone Else Will Be Irreversible
Unlock all features
FREE: Get instant access to 10 AI summaries, chats, or transcripts per day.
Unlock all features
FREE: Get instant access to 10 AI summaries, chats, or transcripts per day.
Unlock all features
FREE: Get instant access to 10 AI summaries, chats, or transcripts per day.
Unlock all features
FREE: Get instant access to 10 AI summaries, chats, or transcripts per day.
Unlock all features
FREE: Get instant access to 10 AI summaries, chats, or transcripts per day.
Related videos
Google VP: The AI Shift Is Done and the Gap Between People Is Growing.Here's How to Stay Ahead
Silicon Valley Girl
36.3k views
LinkedIn Founder: AI Is Changing Every Job Faster Than You Think | Reid Hoffman
Silicon Valley Girl
44.1k views
LinkedIn CEO: These 3 Jobs Will Explode in the Next 5 Years | Ryan Roslansky
Silicon Valley Girl
102.5k views
Godfather of AI: The next 5 years Will Change Humanity Forever | Yoshua Bengio
Silicon Valley Girl
62.1k views
Mel Robbins: How to Get Ahead in 2026 When Everything Feels Uncertain
Silicon Valley Girl
92.2k views
I Tested 100+ AI Tools, These 11 Will Make you Rich in 2026
Silicon Valley Girl
119.5k views
If I need to start over in 2026, Iβd learn these 7 skills first
Silicon Valley Girl
57.5k views
Future of Work 2026: The Only Jobs That Will Survive the AI Era
Silicon Valley Girl
169.8k views
CEO of Microsoft AI:The Next 10 Years Will Change Humanity Forever
Silicon Valley Girl
219.7k views
Anthropic CPO: How AI Will Build the Next $100M Companies | Mike Krieger
Silicon Valley Girl
42.5k views
Top Comments (10)
π What's one task you're still doing manually that you wish you could automate?
Iβm definitely going to have Claude watch the rest of this for me, and send me an email on Friday
Now let's see the difference between the person with the AI, versus the person without the AI, in terms of the gray matter they retain because they're using it regularly versus the gray matter that atrophies because its tasks were offloaded to a machine.
I also have 500 agents working for me right now. Well, 498. Two of them are currently arguing over the best way to fold my socks, finding where I left my keys and the others are just for emotional support
That her first examples of agent workflows is "check industry news," "check on a competitor," and "give me an email summary" is pretty telling.
She has 180 agents running. Now ask her what her monthly token fees are for Claude. lol
hahaha, the intro suggests shes advising anthropic and openai on how to use A!??, thats the funniest thing ive heard in a long time
This could have been 15 mins. Basically- ask claude what you want
I feed the transcript from this conversation into AI and I asked for the blunt feedback on the quality of the conversation and I got this: Person 1 (Alli Miller); Claimed level: Expert; Real level: Intermediate β Advanced-Intermediate. But hereβs the critical part: Most of what she describes is: automation + prompting + connectors; not real βAI engineeringβ. Example: βcomplain to Claude and it builds workflowsβ Thatβs UX-layer AI usage, not deep AI capability. No mention of: model limitations, hallucination mitigation, evaluation loops, fine-tuning, embeddings / retrieval pipelines (RAG), cost optimization, latency tradeoffs π Thatβs a big gap for a βtop AI voiceβ. π¬ Translation into reality: Sheβs very good at: operationalizing AI for business users, packaging workflows, teaching non-technical people. She is NOT: pushing technical boundaries, building novel systems. She use βagentβ loosely. Reality: scheduled prompts, tool calls, API workflows and not: autonomous decision systems.π₯ The uncomfortable truth her job is already: π 50β70% automatable today π 80β90% automatable within ~2 years. Still was good to watch this and was informative.
GREAT episode! SO much amazing content and guidance...a big thanks, to both of you! Just yesterday, I decided to focus on a manual task that I have been doing in Adobe InDesign for the past 10 years. In an afternoon of back & forth, all I need to do now is upload a CSV file, with the customizations (which is the next thing to go), and Claude renders the files almost-exactly as I did manually. "Artistically comparable," I would say, with the correct fonts, and measurements all squared-away.. It's a superpower of-sorts, really...to have a problem, invest some time in working with Claude or Codex or Gemini to help you solve it (I have used all, in the past six months to accomplish a particular stated-goal), and coming out the other side of that invested time with a capability that solves it, in full.
Unlock the Data Inside
Turn Videos into Knowledge
- Get FREE 10/day: transcripts, summaries, chats
- Chat with videos, export text & PDF
- $1 free API credit for RAG, chatbots & research
Free forever plan β’ All features unlocked
Top Comments (10)
π What's one task you're still doing manually that you wish you could automate?
Iβm definitely going to have Claude watch the rest of this for me, and send me an email on Friday
Now let's see the difference between the person with the AI, versus the person without the AI, in terms of the gray matter they retain because they're using it regularly versus the gray matter that atrophies because its tasks were offloaded to a machine.
I also have 500 agents working for me right now. Well, 498. Two of them are currently arguing over the best way to fold my socks, finding where I left my keys and the others are just for emotional support
That her first examples of agent workflows is "check industry news," "check on a competitor," and "give me an email summary" is pretty telling.
She has 180 agents running. Now ask her what her monthly token fees are for Claude. lol
hahaha, the intro suggests shes advising anthropic and openai on how to use A!??, thats the funniest thing ive heard in a long time
This could have been 15 mins. Basically- ask claude what you want
I feed the transcript from this conversation into AI and I asked for the blunt feedback on the quality of the conversation and I got this: Person 1 (Alli Miller); Claimed level: Expert; Real level: Intermediate β Advanced-Intermediate. But hereβs the critical part: Most of what she describes is: automation + prompting + connectors; not real βAI engineeringβ. Example: βcomplain to Claude and it builds workflowsβ Thatβs UX-layer AI usage, not deep AI capability. No mention of: model limitations, hallucination mitigation, evaluation loops, fine-tuning, embeddings / retrieval pipelines (RAG), cost optimization, latency tradeoffs π Thatβs a big gap for a βtop AI voiceβ. π¬ Translation into reality: Sheβs very good at: operationalizing AI for business users, packaging workflows, teaching non-technical people. She is NOT: pushing technical boundaries, building novel systems. She use βagentβ loosely. Reality: scheduled prompts, tool calls, API workflows and not: autonomous decision systems.π₯ The uncomfortable truth her job is already: π 50β70% automatable today π 80β90% automatable within ~2 years. Still was good to watch this and was informative.
GREAT episode! SO much amazing content and guidance...a big thanks, to both of you! Just yesterday, I decided to focus on a manual task that I have been doing in Adobe InDesign for the past 10 years. In an afternoon of back & forth, all I need to do now is upload a CSV file, with the customizations (which is the next thing to go), and Claude renders the files almost-exactly as I did manually. "Artistically comparable," I would say, with the correct fonts, and measurements all squared-away.. It's a superpower of-sorts, really...to have a problem, invest some time in working with Claude or Codex or Gemini to help you solve it (I have used all, in the past six months to accomplish a particular stated-goal), and coming out the other side of that invested time with a capability that solves it, in full.