Open this moment: AgentCDN watch page or YouTube.
by Sander Saar
The discussion introduces how advanced AI models are outpacing current human ambition and business adoption rates. Along the way it covers Why Enterprise AI Deployments Fail and Three Stages of AI Adoption. Shifting roles from doer to orchestrator and applying customer-first principles to unlock AI's full potential.
/public/creators/sander-saar/videos/video_bx8vfR5F06Q/arvp.json
· Browse this video library
The discussion introduces how advanced AI models are outpacing current human ambition and business adoption rates.
On screen: A man addresses the camera against a white background while animated statistical graphic panels appear above him.
Enterprise barriers such as legacy infrastructure, messy data, and security concerns slow down AI implementation.
On screen: A man addresses the camera against a white background while animated statistical graphic panels appear above him.
A breakdown of tech adoption phases—absorb, innovate, and disrupt—and how organizations can move beyond basic automation.
On screen: A man in a brown shirt gestures and explains tech deployment stages directly to the camera against a white backdrop with infographic circles appearing above him.
Shifting roles from doer to orchestrator and applying customer-first principles to unlock AI's full potential.
On screen: A man speaks directly to the camera against a white background with animated infographics appearing above him.
What the footage shows, shot by shot. It is derived from the video itself, independent of the commentary. Timestamps link to the exact second.
0:00–0:08 A man addresses the camera below an animated line graph illustrating the capability overhang between AI models and AI usage. On screen: THE CAPABILITY OVERHANG POWER TIME AI MODELS AI USAGE – Kevin Scott, CTO Microsoft
0:08–0:12 Microsoft CTO Kevin Scott delivers a presentation on stage in front of a slide diagram on agentic AI architectures. On screen: ding the open agentic Agents ry Entitlements Actions Reasoning rotocols (MCP, A2A, etc.) e Web net services Other Kevin Scott Chief Technology Officer…
0:12–0:24 A man in a brown overshirt speaks directly to the camera beneath a benchmark test graphic. On screen: RESEARCH LEVEL AI Crushes Benchmarks Bar Exam Passes in 90th percentile* Writing Tests Exceeds human average Coding Challenges Beats most engineers "Economic…
0:24–0:26 Ilya Sutskever speaks into a microphone in a close-up side-angle interview frame. On screen: Ilya Sutskever Founder Safe Superintelligence SOURCE: DWARKESH PATEL
0:26–0:30 Ilya Sutskever speaks in an outdoor interview setting with a microphone visible in front of him. On screen: Ilya Sutskever Founder Safe Superintelligence SOURCE: DWARKESH PATEL
0:30–0:33 A presenter in an orange-brown overshirt speaks enthusiastically against a white background with animated text. On screen: APPLICATION LEVEL Models are NOT Intelligence Limited They ARE Eval Limited "They can do far more than we're testing them for" - Kevin Weil, OpenAI
0:33–0:35 A blonde woman sits and listens during an interview panel recorded for Lenny's Podcast. On screen: Kevin Weil Chief Product Officer OpenAI SOURCE: LENNY'S PODCAST CONVICTION
0:35–0:41 Kevin Weil sits in a brown armchair on stage, speaking and actively gesturing with his hands during an interview. On screen: Kevin Weil Chief Product Officer OpenAI SOURCE: LENNY'S PODCAST
0:41–1:06 A man addresses the camera against a white background while animated statistical graphic panels appear above him. On screen: USER LEVEL Huge potential. Tiny habits. 12% of Americans use AI daily 25% of CIOs have deployed anything 40% not planning until 2026 "Why did our AI pilot…
1:06–1:11 A man in a brown shirt talks toward the camera beneath animated cloud computing adoption statistics on a white background. On screen: This happened before Cloud Computing 2015 10% adoption 10 years 2025 50% Classic deployment friction takes time
1:12–1:16 The presenter speaks to the camera positioned in front of a ChatGPT app mobile interface graphic. On screen: ChatGPT Can you just write this email for me? Thanks 1 2 3 4 7 8 9 0 #+= Absolutely — here's a clean, tight
1:16–1:17 The man continues explaining tech adoption stages under a title header on a clean white background. On screen: TECH DEPLOYMENT STAGES STAGE 1 OF 3 Absorb Automate obvious use-cases Make it a feature
1:17–1:31 A man in a brown shirt gestures and explains tech deployment stages directly to the camera against a white backdrop with infographic circles appearing above him. On screen: TECH DEPLOYMENT STAGES STAGE 1 OF 3 Absorb Automate obvious use-cases Make it a feature Most companies stuck here STAGE 2 OF 3 Innovate New products, bundling…
1:31–1:34 A man speaks passionately against a white background below a graphic titled 'Innovate'. On screen: TECH DEPLOYMENT STAGES STAGE 2 OF 3 Innovate New products, bundling and unbundling Ship something that only works because AI exists
1:34–1:39 The speaker gestures dynamically under a graphic titled 'Disrupt' on a clean white background. On screen: TECH DEPLOYMENT STAGES STAGE 3 OF 3 Disrupt Redefine the question Build as if starting from zero
1:39–1:48 A man speaks directly to the camera against a white background with animated infographics appearing above him. On screen: The Real Opportunity Doing your job faster Doing a different job Engineer Eng Manager Writer Editor Doer Orchestrator
1:48–1:49 A man gestures while an animated gauge graphic dials up to eleven above his head on a white background. On screen: 0 3 6 9 11 SET AMBITION TO
1:49–1:52 Kevin Scott speaks into a microphone while presenting on stage in front of a blue background graphic. On screen: Int Kevin Scott Chief Technology Officer Microsoft SOURCE: MICROSOFT DEVELOPERS
1:52–1:55 Kevin Scott presents on stage holding a remote clicker while gesturing with his hands. On screen: Kevin Scott Chief Technology Officer Microsoft SOURCE: MICROSOFT DEVELOPERS
1:55–1:57 The presenter speaks to the camera beneath an on-screen graphic referencing Steve Jobs' approach. On screen: START WITH Experience Work backwards to Technology Steve Jobs
1:57–2:03 Steve Jobs delivers a speech on stage, gesturing expressively with both hands. On screen: Steve Jobs Co-Founder Apple SOURCE: JONATHAN FIELD
2:03–2:09 Archival footage shows seated audience members listening attentively in a darkened auditorium while text identifies Steve Jobs speaking. On screen: Steve Jobs Co-Founder Apple SOURCE: JONATHAN FIELD
2:09–2:11 A man in a brown shirt speaks to the camera against a plain white background beneath an animated gauge graphic. On screen: The models are ready. We're the ones lagging. 0 11 AMBITION TO
0:00 Speaker A: We are massively underusing today's AI, not because the models are weak, but because our ambition is Microsoft CTO Kevin Scott calls this capability overhang. Speaker B: So the models are more powerful than what we collectively are using them for. Speaker A: This overhang exists at every level at the research and model level. Speaker A: Ilya Sutskeverg, the co founder of OpenAI, points out that models now crush hard benchmarks. Speaker A: They're surpassing humans in law with bar exam competitive coding, data analysis. Speaker C: They are doing so well on evals, but the economic impact seems to be dramatically behind at the application level.
0:31 Speaker A: Kevin Wilde, chief product officer of OpenAI,. Speaker D: Says models today are not intelligence limited, they're eval limited. Speaker D: They can actually do much more and be much more correct on a wider range of things than they are today. Speaker A: And at the user level. Speaker A: While about half of the Americans have tried AI, only about 1 in 8 less than 12% use it daily in businesses. Speaker A: Only 25% of the CIOs have deployed anything significant and 40% aren't planning it until 2026. Speaker A: Why did our AI pilot fail? Speaker A: Or your friend telling AI doesn't work?
0:57 Speaker A: It's almost never the AI, it's the CTO and the product question the legacy systems, messy data security, privacy worries that still consume most of the tech budget. Speaker A: Classic tech deployment problem. Speaker A: The same reason why cloud still runs only about a third to half of the enterprise workloads 10 years later. Speaker A: And when we do use AI, most of us just do the same job a bit faster. Speaker A: Ben Evans showed that tech adoption in three stages. Speaker A: First absorb, automate the obvious stuff, type in a prompt, get a paragraph and done. Speaker A: Marshall McLuhan called this the horse's carriage syndrome.
1:23 Speaker A: Using advanced tech for the most basic things. Speaker A: It's useful but it's limited. Speaker A: But this is where most teams are. Speaker A: Next level is innovation means shipping something that only works because AI exists. Speaker A: This is where we start unlocking true value. Speaker A: Finally disrupt build like you started from zero. Speaker A: If you rebuilt the product from scratch, what would it look like? Speaker A: The real opportunity isn't doing your job faster, it's doing a different job entirely. Speaker A: Moving from doer to orchestrator, engineering to engineering manager writer to editor Kevin Scott said the challenge Think about how you.
1:50 Speaker B: Can set your ambition level to 11 target some things that are you think are barely possible. Speaker A: And Steve Jobs gave us the playbook for new tech. Speaker B: You can't start with the technology and and try to figure out where you're going to try to sell it. Speaker B: You've got to start with the customer experience and work backwards to the technology. Speaker A: The models are ready. Speaker A: We're the ones lagging.