"one day early morning President Trump asking me to resign and conflict of interest and there's no exceptions... I had to convince myself first of all you know I don't need this job I do it purely to save Intel." - Lip Bu Tan [00:02:02]
"first of all you crawl and then be humble listen to customer and then secondly deciding to walk and then finally deciding to run in spring so that's kind of my culture of step by step doing it" - Lip Bu Tan [00:04:14]
Disclaimer: Orignal content owned by or sourced from third parties. It does not represent the views of 'Nuggets' platform or it's team. AI is used extensively across this platform including for summaries. Accuracy is not guaranteed, there can be mistakes. Any info or content on this platform is not a financial, legal, or investment advice. Do your own research. Refer for complete disclosures:- Terms of Use · Full Disclaimer
"one thing good about being an engineers you always hitting the wall then you find way to either jump over the wall or you work around the wall and then uh to get to the better result" - Lip Bu Tan [00:17:49]
"initially the partners meeting the whole partners in the room then after I talking about semiconductor half make excuse to run out of the room then eventually the other half they said available do you have any software service so then they left with only who sympathetically listen to me" - Lip Bu Tan [00:19:24]
"nine of the 10 company I invest halfway they change their business plan because market have changed so I like to have entrepreneur as team not just one person" - Lip Bu Tan [00:30:13]
"so you know I always look for 10x you know being adventure at heart you want to look for 10x... and five year 10 years if we can do 10x I think is a good return" - Lip Bu Tan [00:40:25]
Speakers & Credentials
Sarah Guo: Host of No Priors, Co-founder of Conviction, leading investor in AI and frontier technology.
Elad Gil: Host of No Priors, renowned Silicon Valley entrepreneur, operator, and prolific early-stage investor.
Lip Bu Tan: CEO of Intel (executing a historic corporate turnaround), former CEO and Executive Chairman of Cadence Design Systems (where he drove an 85x stock return over 15 years), and Founder/Chairman of Walden International, an elite venture capital firm with an iconic track record in semiconductor investing (159 IPOs, 126 M&A exits).
1. Executive Summary
At 66 years old, legendary investor and executive Lip Bu Tan took the helm as CEO of Intel, motivated by an industry-critical mission to revive the iconic American semiconductor giant.
He is aggressively stripping away Intel's legacy bureaucracy, enforcing a flat engineering reporting structure directly to himself, and instituting a strict "crawl, walk, run" operational strategy starting with core balance sheet stabilization.
Tan views domestic semiconductor manufacturing as an existential imperative for supply chain resilience, doubling down on the capital-intensive foundry business to provide a secure alternative to TSMC.
To overcome the looming physical limits of Moore's Law, Intel is pivoting heavily into advanced material sciences—investing in glass substrates, artificial diamonds, silicon carbide, and gallium nitride to solve complex power, thermal, and interconnect bottlenecks.
Drawing from his unparalleled venture capital track record (159 IPOs), Tan believes the future of the industry belongs to "full-stack" platforms, predicting that agentic and physical AI will spark a massive shift toward highly specialized, workload-specific edge silicon.
2. Chronological Table of Contents
[00:00:00] Introduction & Why Take the Intel Job at Age 66
[00:02:02] Overcoming Government Friction: The Trump Ultimatum
[00:03:01] Re-engineering Intel's Culture: Speed, Accountability, and Flat Reporting
[00:04:14] The Crawl, Walk, Run Strategy & Balance Sheet Strengthening
[00:06:17] The Economics and Mechanics of the Foundry Business
[00:08:00] Unconventional Partnerships: Building "Terafab" with Elon Musk
[00:10:01] Geopolitics, Global Supply Chains, and Infrastructure Constraints
[00:15:35] Escaping the Limits of Physics: Next-Generation Material Science
[00:18:42] Thematic History: The 20-Year Desert of Semiconductor Venture Capital
[00:24:06] Bottleneck Investing: Power, Thermal, and Connectivity
[00:28:06] The Next 10 Years: Niche Dominance and Full-Stack Solutions
[00:32:05] Organizational Transformation: Managing Agents and Aging Workforces
[00:34:35] Sovereign Wealth, Industrial Policy, and Massive Valuations
[00:37:36] Intel's 10x Horizon and Edge Compute Realities
3. Detailed Thematic Summary
The Existential Turnaround of an American Institution
Upon taking the CEO position at age 66, Lip Bu Tan faced immediate governmental friction when President Donald Trump asked him to resign due to conflict-of-interest concerns [00:02:02]. Tan leveraged his personal history—born in Malaysia, raised in Singapore, educated at MIT, and a lifelong US resident—to convince the administration his only goal was to "save Intel" [00:02:44].
To dismantle a sluggish legacy bureaucracy that operated via "layers of meetings," Tan stripped the hierarchy so that all engineering now reports directly to him, ensuring he can intimately understand technical failures and accelerate decision-making [00:03:32].
His first phase ("Crawl") focused on rescuing Intel's "horrible" balance sheet, securing vital support from the US government, a $5 billion investment from his longtime friend Jensen Huang (which has since multiplied to $25 billion), and assistance from SoftBank's Masayoshi Son [00:04:33].
Tan's ultimate goal for Intel is a venture-like return; having generated an 85x return over 15 years at Cadence (growing the stock from an interim $2.42), he is targeting a 10x return for Intel shareholders over the next 5 to 10 years, expecting the fruits of this labor to fully surface between 2030 and 2032 [00:39:09], [00:40:25].
Reshaping the Global Foundry & Manufacturing Paradigm
Despite heavy market skepticism about the exorbitant capital requirements, Tan decided to double down on Intel's foundry business, citing it as critical for the resilience of the United States' supply chain, as dependence on a single geographic node (like TSMC in Taiwan) is an unacceptable macro risk [00:13:52].
He acknowledges that Intel is currently "very distant" from TSMC in performance, noting that the foundry model is fundamentally a "trust business" requiring precise execution on yield, defect density, and cycle times [00:38:40].
Intel is collaborating with Elon Musk on "Terafab," an unconventional effort to build a factory at an accelerated pace, questioning traditional clean-room orthodoxies to rapidly scale silicon production for Musk's AI and robotics ecosystem [00:08:00].
The scale of precision is staggering: Intel is currently producing the 18A node, moving into 14A (1.4 nanometer), and actively planning for 1 nanometer and 0.7 nanometer processes—widths comparable to a single hair, where a single misstep sends millions of dollars "down the drain" [00:14:43].
The Physics of Compute & The Next Era of Materials
As traditional silicon approaches terminal physical limits regarding line width and heat generation, Tan's investment and product strategy is pivoting heavily toward exotic material science and advanced packaging to "jump over the wall" [00:17:49].
The industry bottleneck is shifting from pure compute density to interconnect speed, power conversion, and thermal management; converting 40 volts down to 1 volt safely requires immense innovation [00:24:25].
Tan is actively investing in new substrate materials, heavily targeting gallium nitride, silicon carbide, and indium phosphide to manage power [00:16:43].
For advanced packaging (competing with TSMC's CoWoS), he is exploring the thermal insulation properties of glass (via investments in 3DGS) and artificial diamonds (investing in Diamond Foundry), viewing these as the next generation of semiconductor infrastructure [00:17:04].
Thematic History: The 20-Year Desert of Semiconductor Venture Capital
To understand the current AI boom, Tan contextualizes the historical irrelevance of hardware to venture capital. Eighteen years ago, when pitching semiconductor investments, half the VC partners would literally make excuses to run out of the boardroom, while the remaining half would beg him to pivot to SaaS software [00:19:24].
Because hardware was viewed as cyclical, unpredictable, and highly capital intensive, Tan operated with a tiny coalition of strategic corporate backers like Samsung, ARM, and SoftBank [00:20:16].
This deep-time context highlights the irony and whiplash of the modern market, where formerly ignored hardware companies now command sovereign-level valuations: Nvidia at $5.3 Trillion, TSMC and Broadcom at $2 Trillion, AMD near $800 Billion, and Intel at $600 Billion [00:19:52].
Operating through this desert built Tan's legendary venture track record: generating 159 IPOs and 126 M&A exits, keeping a remarkably high 38% of those investments domiciled in the United States [00:22:42].
The Agentic Future and the Edge Compute Reality
The macro landscape of AI workloads is shifting violently. Tan notes that the ratio of training GPUs to inference CPUs was previously 1:8, but the rise of agentic AI is driving that ratio to 1:4, creating a massive resurgence in demand for Intel's core CPU products [00:05:34].
He predicts a divergence in compute architecture: while massive data centers continue to grow, the rise of "physical AI" (robotics) and millions of autonomous software agents will necessitate massive, highly efficient edge compute closer to the client [00:39:30].
Internally, Intel is undergoing a demographic and operational shift, moving away from an aging workforce (average age late 40s/50s) overly reliant on legacy spreadsheets, recruiting younger software-native talent to integrate open-source and frontier models to transform Intel into an AI-enabled design firm [00:33:18].
The Reference Vault
4. Data & Figures
Data Point
Value
Context
Timestamp
Lip Bu Tan's Age
66
The age at which he stepped out of retirement to become CEO of Intel.
The "Crawl, Walk, Run" Restructuring Heuristic [00:04:14]
Lip Bu Tan applies this sequential framework to corporate turnarounds to prevent organizational shock. "Crawling" dictates severe humility—listening to neglected customers and performing triage on bleeding balance sheets before launching flashy products. "Walking" involves pruning bloated product lines and establishing clear, simplified roadmaps. Only once operational trust is restored does a company "Run," attempting to sprint toward full-stack integration and next-generation architectures. This framework acts as an ego-check, preventing failing giants from trying to innovate before fixing basic foundational leaks.
"Bottleneck Investing" (The Constraint Theory of Venture) [00:24:06]
Rather than investing in generalized technology trends, Tan isolates the exact physical or systemic chokepoint threatening a macro-trend and aggressively funds the solutions targeting it. As AI scales, raw compute is no longer the bottleneck—power, thermal management, and data transfer speeds are. By systematically backing material sciences (gallium nitride for power reduction, artificial diamonds for heat insulation, silicon photonics for data transfer), Tan turns the inherent constraints of physics into his primary investment thesis.
The "Hyperscaler First" Go-To-Market Protocol [00:24:56]
Hardware start-ups face a devastating "valley of death" due to immense capital requirements and customer switching costs. Tan mitigates this by mandating that founders target a hyperscaler (e.g., AWS, Microsoft, Google) as their "Customer Zero" from day one. If a hyperscaler commits to a customized solution, they bring not only millions in immediate capital but also unparalleled market validation. Conceding warrants or deep equity to these giants is deemed strategically necessary because securing a single hyperscaler effectively guarantees the startup will survive to scale.
The Imperative of the "Full Stack" Platform [00:31:01]
In the modern semiconductor landscape, isolated silicon is commoditized; ecosystems win. Tan points to Nvidia's CUDA as the archetypal full-stack moat. To survive the next decade, semiconductor companies cannot simply produce chips; they must provide integrated software layers, orchestration tooling, and advanced packaging to deliver "the whole rack." Intel’s survival hinges not just on reclaiming process leadership, but on merging XPU processing, foundry services, and software architecture into a singular, frictionless platform for physical AI developers.
6. Anecdotes
The Presidential Ultimatum [00:02:02]
Context & Meaning: Shortly after stepping in to help rescue Intel, Tan was confronted by President Donald Trump early in the morning and ordered to resign due to "conflict of interest." Tan shared this to illustrate the profound, unpredictable political friction involved in leading a structurally critical American company. Instead of backing down, Tan successfully managed the encounter by appealing to his immigrant background, MIT education, and pure intention to save a US institution, highlighting his diplomatic resilience in navigating geopolitical pressure.
The 3-Month Commitment that Turned into 15 Years [00:07:39]
Context & Meaning: When Tan initially took over Cadence, he signed on as an interim CEO strictly for "three months." That temporary stint transformed into a legendary 15-year tenure that drove the company's stock up 85x. He recounted this to humorously warn against the danger of temporary commitments, but strategically, it signals his deep stamina and long-term commitment. It suggests to investors that his current role at Intel, while undertaken later in life, is backed by a track record of enduring execution, not short-term tinkering.
The Empty VC Boardroom of the Mid-2000s [00:19:24]
Context & Meaning: Eighteen years ago, when Tan pitched semiconductor startups to top-tier Sand Hill Road venture firms, partners would literally fabricate excuses to leave the room, viewing hardware as a toxic, un-fundable sector compared to SaaS software. Tan told this story to highlight the intense irony of the current market, where the very hardware companies VCs used to shun now command the highest market capitalizations in human history (Nvidia at $5.3 Trillion). It proves the value of contrarian conviction.
Learning AI from his Son [00:33:18]
Context & Meaning: Despite his elite status as a hardware titan, Tan openly admits that he relies on his son as his primary teacher for modern open-source frontier models and software-agent dynamics. He shared this anecdote to underscore the radical demographic and educational shift required at Intel. Recognizing his own generational blind spots, he is using this mindset to update Intel's talent pool, transitioning the company from an aging, spreadsheet-driven culture to a younger, AI-native engineering force.
7. References & Recommendations
People
Donald Trump: Former US President, who initially asked Tan to resign over conflict-of-interest concerns before Tan successfully lobbied him to stay and execute the Intel turnaround. [00:02:02]
Jensen Huang: CEO of Nvidia and old friend of Tan; invested an initial $5B in Tan/Intel which blossomed into $25B, cited as an example of vital industry partnership and asset appreciation. [00:04:47]
Masayoshi Son (Masa): Founder of SoftBank; cited by Tan as a vital capital partner who helped lend a hand to strengthen Intel's balance sheet during the triage phase. [00:05:13]
Elon Musk: CEO of Tesla/SpaceX; currently collaborating with Intel to build "Terafab," pushing conventional fab rules to accelerate custom silicon for his robotics efforts. [00:08:00]
Lisa Su: CEO of AMD; praised by Tan as a "good friend," acknowledging her massive success in driving AMD to an $800 Billion market cap. [00:19:52]
Sassine Ghazi: CEO of Synopsys; mentioned by Tan as a successor generation driving the implementation of agentic AI into EDA tooling. [00:27:05]
Companies & Institutions
Intel: The iconic American semiconductor firm Tan is currently restructuring to ensure global supply chain dominance. [00:01:32]
Cadence Design Systems: The massive EDA software company Tan ran for 15 years, orchestrating a monumental 85x turnaround. [00:04:14]
Synopsys: Major EDA software competitor mentioned in the context of leveraging AI to automate chip layouts and design flows. [00:27:05]
Ansys: Engineering simulation software company acquired by Synopsys to expand full-system simulation capabilities. [00:27:12]
Siemens: Industrial conglomerate mentioned as a major software/hardware acquirer of niche electronic design startups. [00:27:24]
Walden International: Tan's elite venture capital firm responsible for 159 semiconductor IPOs. [00:22:42]
TSMC (Taiwan Semiconductor Manufacturing Company): The reigning champion of global foundries; cited heavily as the primary benchmark and competitor Intel is attempting to catch up to. [00:13:52]
Nvidia: The $5.3 Trillion AI behemoth; cited as the paragon of full-stack platform building (CUDA) that Intel must emulate. [00:19:52]
Cradle Semiconductor, Celestial AI, Inphi: Portfolio companies Tan invested in to solve optical interconnect and advanced cluster data speed bottlenecks. [00:24:06]
Analog Devices (ADI) / Empower: Mentioned in the context of ADI acquiring Empower to mitigate the extreme power losses that occur when converting 40V down to 1V. [00:24:25]
3DGS & Diamond Foundry: Next-generation material science companies Tan is backing to solve heat constraints in advanced packaging using glass and artificial diamond. [00:17:04]
Anthropic & OpenAI: Frontier AI labs cited as examples of aggressive, nimble start-ups changing the fundamental nature of compute demand. [00:31:35]
Periodic: A specialized materials company highlighted as part of the next-generation wave altering traditional manufacturing. [00:28:06]
Geopolitics & Macro-Ecosystems
US Government / Sovereign Funds: Essential capital partners required to underwrite the immense infrastructure cost of repatriating semiconductor foundries. [00:34:35]
Indian Government: Partnering with Intel on a major advanced packaging program to diversify global infrastructure footprint. [00:17:18]
Israel: Highlighted by Tan as a wildly resilient tech ecosystem, noting that Israeli founders continue to pitch and build seamlessly even while sheltering underground during wartime. [00:25:34]
8. The Bottomline (by AI)
The era of cheap, centralized, unconstrained silicon is over; the future of compute is a brutal war over material physics, thermal dynamics, and massive sovereign capital allocation. Intel's turnaround is not just a corporate restructuring, but a geopolitical imperative to secure the Western hemisphere's physical supply chain against structural single-point vulnerabilities. Investors and technologists must watch the shift from traditional silicon scaling to exotic materials (glass, diamonds, silicon carbide) and monitor how the explosive rise of agentic and physical AI forces a resurgence of localized edge computing over pure cloud dominance. Survival in this decade demands a "full-stack" platform approach, turning custom hardware and software ecosystems into impenetrable moats.
Jul 16, 2026
Dr. Robert Wachter | A Giant Leap: How AI Is Transforming Healthcare... | 14 Jul 2026 | Talks at Google
"don't get me wrong US healthcare delivers miracles every day particularly when it comes to cutting edge and intensive care... but the health care system itself is a headache wrapped in red tape inside the nightmare that France Kofka himse…
Intel Target Nodes
18A, 14A, 1nm, 0.7nm
The specific semiconductor manufacturing milestones Intel is actively pushing toward.