Generating Alpha Podcast
1 hr 15 min video
3 min read
Gavin Baker on AI, Investing, and Building Conviction
You just saved 1 hr 12 min.
The big takeaway
Gavin Baker, managing partner of Atreides Management and former Fidelity tech portfolio manager, discusses his journey from semiconductors to AI investing, the importance of resilience through market downturns, and how finding an investment philosophy that matches your emotional makeup is key to long-term success. He argues AI is not in a valuation bubble but may face a capex bubble, emphasizes the role of semiconductors in AI's future, and shares hard-won lessons on handling career setbacks.
Early Life and Formation
Free-range intellectual childhood shaped curiosity
Baker grew up in Texas with parents who encouraged unlimited reading and exploration. He spent summers in the Hill Country with cousins, had an unlimited book budget, and was exposed to history, games of skill, and outdoor activities. This environment fostered his love of learning and pattern recognition.
Parents' strategic constraints led to investing career
Baker's parents made only two requests: no fraternity in freshman year, and one real internship before pursuing his original plan of climbing and writing. The internship at Donaldson, Lufkin & Jenrette reading research reports sparked his realization that investing was a game of skill and chance combining history with differential opinion.
Fidelity as the ultimate training ground
Baker joined Fidelity in 1999 as an analyst and credits it as the best place to learn investing. Fidelity exposed him to every investment philosophy imaginable, allowed him to learn from legendary PMs managing billions, and gave him freedom to generate alpha through outcome-based work rather than rigid process.
Career Turning Points and Resilience
Dot-com bubble analysis launched semiconductor expertise
In February 2000, Baker published a note analyzing semiconductor and customer inventory levels at all-time highs despite sky-high valuations. This analysis, which proved prescient as the bubble burst, caught the attention of senior analyst Rajiv Call and established Baker's credibility in semiconductors.
Feb 2000
Baker publishes inventory analysis note
2000
Dot-com bubble peaks and crashes
Post-2000
Baker becomes recognized semiconductor expert
Baker's prescient 2000 analysis during the dot-com peak
Early career humbling on pharma fund
At age 25, Baker was promoted to manage a large-cap pharma fund but got the sector call wrong, dropping from top-ranked analyst to bottom decile. This painful experience taught him that resilience and loving the game are more important than early success, and that going through hardship early in your career is advantageous.
Small-cap semiconductors
Top-ranked analyst
Large-cap pharma
Bottom decile performer
Baker's career inflection point at age 25
Touchstones and habits for navigating drawdowns
Baker developed coping mechanisms for tough periods: rereading 'A Wizard of Earthsea' as a reminder about arrogance, exercising to create measurable improvement, reviewing past journal entries, and remembering that being in the bottom decile is statistically as difficult as being in the top decile. These practices helped him maintain decision quality during volatility.
Litigation expertise in tobacco sector
After pharma, Baker covered tobacco and had to call three existential litigation cases correctly or face termination. He attended every trial and hearing, consulted with former judicial clerks, and read tens of thousands of pages to gain an edge in binary outcomes. This experience taught him that deep work creates alpha even in uncertain situations.
Semiconductor and Tech Investing Philosophy
Semiconductors are the foundation of AI's future
Baker argues that AI is fundamentally different from software because it is probabilistic and computationally expensive, requiring recomputation each time. This makes semiconductors—watts and wafers—essential infrastructure. His deep tech background in semiconductors positioned him perfectly for the AI era.
Tesla's first-principles EV advantages
Baker attended a 6 p.m. Tesla meeting when the company had a $1.5 billion market cap. Elon explained that EVs are superior because battery costs are deflationary (unlike inflationary engine components), batteries enable lower center of gravity and better handling, electric motors provide millisecond traction control, and the absence of a large engine block allows dual crumple zones for safety.
1
Battery cost trajectory
Deflationary (mid-to-high single digits annually)
2
Engine components
Inflationary
3
EV handling advantage
Lower center of gravity + lower polar moment of inertia
4
Safety advantage
Dual crumple zones vs. single engine block
Tesla's structural advantages from first principles
25 years following Nvidia closely
Baker has read every Nvidia transcript and followed the company for 25 years, even when not covering it. He reads GPU head-to-head comparisons and games with high-end GPUs to understand the product. This long-term, close observation of exceptional companies is central to his investing approach.
AI, Valuations, and Market Bubbles
History shows revolutionary tech always creates bubbles
Baker cites Carlota Perez's 'Financial Capital and Technological Revolutions,' which shows that canals, railroads, radio, the internet, and PCs all created bubbles. A bubble occurs when financial markets correctly identify revolutionary technology but then experience a breakdown in diversity of opinion, leading to overbuild and crash.
1800s
Canal and railroad bubbles
1920s
Radio bubble
1990s-2000
Internet and PC bubble
2020s
AI era—capex bubble risk?
Historical pattern of bubbles following revolutionary tech
AI not in valuation bubble but may face capex bubble
Baker argues tech stocks trade at the same multiples as five years ago and are now at a discount to staples—not a valuation bubble. However, the massive capex spend on AI infrastructure raises the question of whether there is a capex bubble. So far, ROI on AI spending has been positive, though currently in a dip due to Blackwell training costs.
Same multiple as 5-6 years ago
Tech valuation level
Tech valuations have not inflated despite AI hype
Watts, wafers, and scars prevent AI overbuild
Baker identifies three factors that may prevent an AI capex bubble: (1) fundamental shortage of watts and wafers needed for data centers, (2) TSMC's limited willingness to expand capacity, and (3) deep scars from the 2008 financial crisis that keep a lid on tech valuations. These constraints may naturally limit overbuild.
1
Watts and wafers shortage
Physical constraint
2
TSMC capacity expansion
Limited by management choice
3
2008 financial crisis scars
Psychological constraint on valuations
Three factors limiting AI capex bubble risk
Tech multiples compressed since 2025 start
Baker notes that tech multiples have actually compressed since the beginning of 2025, and tech now trades at a discount to staples—a rare occurrence. This suggests the market is not pricing in excessive AI optimism.
AI, Creativity, and the Future of Work
Human creativity may not be irreplaceable
Baker acknowledges Elon's concept that humans may be the biological bootloader for digital superintelligence. He suggests that human creativity may not be irreplaceable, and that finding new sources of meaning beyond work will be important as AI advances.
AI will transform video game development economics
AI world models will reduce AAA game development costs by 90%, from $100-300 million to much lower levels. This will increase competition among game developers but benefit video game platforms through an explosion of content. However, rendering will remain local on GPUs for the foreseeable future due to computational costs.
AAA game development cost
$100-300 million
With AI world models
90% reduction
AI's impact on game development economics
Neuralink as human-AI interface solution
Baker sees Neuralink as addressing humanity's IO problem—the limitation of speech, writing, and typing. By enabling faster information input and output, Neuralink could help biological intelligences coexist with digital intelligences and maintain value in an AI-dominated world.
Human brains are energy-efficient compared to AI
The human brain operates on 20-30 watts while training AI models requires hundreds of megawatts and inference requires far more than 20-30 watts. In a watt and energy-constrained world, Baker is optimistic that human intelligence will retain value for a long time.
Human brain (inference)
25 watts
AI model training
500 megawatts
AI model inference
100 watts (estimated)
Energy efficiency: human brain vs. AI systems
Investment Philosophy and Mindset
Find philosophy and process that fit your emotional makeup
Baker emphasizes that successful investing requires finding an investment philosophy and process that aligns with your unique emotional makeup so you can remain rational when wrong. Most investors start as value investors because they've read Graham and Buffett, but this may not suit everyone's temperament.
Being early is the same as being wrong
Baker quotes Fidelity PM George Vanderheiden: 'Being early is the same thing as being wrong.' If you're losing money, you're wrong—not the market, not other people. This mindset forces accountability and clear decision-making during drawdowns.
Either panic early or double down late
Baker's mentor Jennifer Yuric taught him that as an investor, you must either panic early or double down late—almost no one does both. Knowing yourself means understanding which type you are. Baker identifies as a 'double down late' person, which shapes how he navigates tough periods.
Hardship early in career is advantageous
Baker argues that going through a major drawdown or failure early (he experienced this at age 25 on pharma) is actually advantageous because you learn resilience when the stakes are lower. The earlier you encounter hardship, the better you navigate it later when consequences are higher.
Decision quality improves after multiple drawdowns
Baker's goal is to reach a point where decision quality actually improves during drawdowns because he has data from past difficult periods. Instead of being gun-shy, he becomes more confident because he has proven he can make high-quality decisions under pressure.
Advice for Young Investors and Professionals
Be kind and be scrappy
Baker's core advice is to be both kind and scrappy—most people excel at one but not both. Kindness creates karma and attracts like-minded people who cooperate. Being scrappy means persistence and hustle. Together, they create a network of mutual support.
Your reputation is made in your early 20s
Baker cites Steve Schwarzman's biography: people remember how you behaved in your early career, not just your performance. Relationships formed early can resurface 20 years later when expertise is needed. This makes treating peers well a strategic investment.
Persistence and value delivery unlock mentorship
To get mentored by someone you don't know, Baker recommends: (1) be persistent and don't give up quickly, and (2) deliver value before asking for value. Try to know something the mentor doesn't, or offer insights they can use.
Only be an investor if you love the game
Baker emphasizes that public markets investing requires loving the game because you will go through really tough periods. If you can't find emotional and psychological centeredness during your first major drawdown, consider private equity or other paths with less volatility.
Seek hardship early in your career
Baker references the Sun Eater sci-fi series mantra: 'Seek hardship.' No matter how good you are, you will encounter it. The decisions you make during those times will define your career. Better to encounter it early when stakes are lower.
Elon and Mission-Driven Companies
Mission-oriented companies attract exceptional talent
Baker argues that Elon's success stems partly from creating mission-oriented companies. Talented engineers can choose between working on incremental ad clicks for Google/Meta or decarbonizing the world (Tesla), making humanity multi-planetary (SpaceX), or enabling free speech (X). Authentic missions attract exceptional people.
Tesla accelerated EV adoption by 20-30 years
Baker credits Tesla and Elon with accelerating EV adoption by 20-30 years and doing more for decarbonization than all environmental activists combined. This acceleration has had massive positive impact on the environment and the world.
20-30 years
EV adoption acceleration by Tesla
Tesla's impact on the timeline of EV adoption
Historians will view this as the age of Elon
Baker believes historians will look back on this era as the age of Elon because of his continued work on multiple world-changing missions (Tesla, SpaceX, Neuralink, X) while most centibillionaires have stopped working. Without Elon's continued effort, there would be no Starlink and lower odds of seeing a blue sunset on Mars.
Worth quoting
"Being early is the same thing as being wrong. If you're losing money, you're wrong."
— Gavin Baker, at [63:23]
"You either have to panic early or double down late. And essentially, no one does both."
— Jennifer Yuric (via Gavin Baker), at [63:54]
"Seek hardship. The decisions you make during those times will define your career."
— Gavin Baker, at [73:08]
Try this
Develop a touchstone (book, movie, or practice) to return to during market drawdowns and tough periods
Create a hobby or measurable goal (running, weightlifting, gaming) that lets you improve a number during difficult performance periods
Document your investment philosophy and process in writing, aligned with your emotional makeup, before you need it
Review past journal entries and decisions from previous drawdowns to build confidence in your decision-making under pressure
Practice being kind and scrappy simultaneously: build a network through genuine relationships while maintaining persistence in pursuing opportunities
When seeking mentorship, deliver value first before asking for it—share insights or knowledge the mentor doesn't have
Read 'A Wizard of Earthsea' and 'The Perfect Storm' as touchstones for navigating arrogance and perseverance through hardship
Made with Glimpse by Wozart
glimpse.wozart.com/v/qk58hfe3
Share this infographic
Read this infographic as text

Gavin Baker on AI, Investing, and Building Conviction

Summary of the video “Episode 56: Gavin Baker - Managing Partner of Atreides Management by Generating Alpha Podcast.

Gavin Baker, managing partner of Atreides Management and former Fidelity tech portfolio manager, discusses his journey from semiconductors to AI investing, the importance of resilience through market downturns, and how finding an investment philosophy that matches your emotional makeup is key to long-term success. He argues AI is not in a valuation bubble but may face a capex bubble, emphasizes the role of semiconductors in AI's future, and shares hard-won lessons on handling career setbacks.

Early Life and Formation

Free-range intellectual childhood shaped curiosity

Baker grew up in Texas with parents who encouraged unlimited reading and exploration. He spent summers in the Hill Country with cousins, had an unlimited book budget, and was exposed to history, games of skill, and outdoor activities. This environment fostered his love of learning and pattern recognition.

Parents' strategic constraints led to investing career

Baker's parents made only two requests: no fraternity in freshman year, and one real internship before pursuing his original plan of climbing and writing. The internship at Donaldson, Lufkin & Jenrette reading research reports sparked his realization that investing was a game of skill and chance combining history with differential opinion.

Fidelity as the ultimate training ground

Baker joined Fidelity in 1999 as an analyst and credits it as the best place to learn investing. Fidelity exposed him to every investment philosophy imaginable, allowed him to learn from legendary PMs managing billions, and gave him freedom to generate alpha through outcome-based work rather than rigid process.

Career Turning Points and Resilience

Dot-com bubble analysis launched semiconductor expertise

In February 2000, Baker published a note analyzing semiconductor and customer inventory levels at all-time highs despite sky-high valuations. This analysis, which proved prescient as the bubble burst, caught the attention of senior analyst Rajiv Call and established Baker's credibility in semiconductors.

Early career humbling on pharma fund

At age 25, Baker was promoted to manage a large-cap pharma fund but got the sector call wrong, dropping from top-ranked analyst to bottom decile. This painful experience taught him that resilience and loving the game are more important than early success, and that going through hardship early in your career is advantageous.

Touchstones and habits for navigating drawdowns

Baker developed coping mechanisms for tough periods: rereading 'A Wizard of Earthsea' as a reminder about arrogance, exercising to create measurable improvement, reviewing past journal entries, and remembering that being in the bottom decile is statistically as difficult as being in the top decile. These practices helped him maintain decision quality during volatility.

Litigation expertise in tobacco sector

After pharma, Baker covered tobacco and had to call three existential litigation cases correctly or face termination. He attended every trial and hearing, consulted with former judicial clerks, and read tens of thousands of pages to gain an edge in binary outcomes. This experience taught him that deep work creates alpha even in uncertain situations.

Semiconductor and Tech Investing Philosophy

Semiconductors are the foundation of AI's future

Baker argues that AI is fundamentally different from software because it is probabilistic and computationally expensive, requiring recomputation each time. This makes semiconductors—watts and wafers—essential infrastructure. His deep tech background in semiconductors positioned him perfectly for the AI era.

Tesla's first-principles EV advantages

Baker attended a 6 p.m. Tesla meeting when the company had a $1.5 billion market cap. Elon explained that EVs are superior because battery costs are deflationary (unlike inflationary engine components), batteries enable lower center of gravity and better handling, electric motors provide millisecond traction control, and the absence of a large engine block allows dual crumple zones for safety.

25 years following Nvidia closely

Baker has read every Nvidia transcript and followed the company for 25 years, even when not covering it. He reads GPU head-to-head comparisons and games with high-end GPUs to understand the product. This long-term, close observation of exceptional companies is central to his investing approach.

AI, Valuations, and Market Bubbles

History shows revolutionary tech always creates bubbles

Baker cites Carlota Perez's 'Financial Capital and Technological Revolutions,' which shows that canals, railroads, radio, the internet, and PCs all created bubbles. A bubble occurs when financial markets correctly identify revolutionary technology but then experience a breakdown in diversity of opinion, leading to overbuild and crash.

AI not in valuation bubble but may face capex bubble

Baker argues tech stocks trade at the same multiples as five years ago and are now at a discount to staples—not a valuation bubble. However, the massive capex spend on AI infrastructure raises the question of whether there is a capex bubble. So far, ROI on AI spending has been positive, though currently in a dip due to Blackwell training costs.

Watts, wafers, and scars prevent AI overbuild

Baker identifies three factors that may prevent an AI capex bubble: (1) fundamental shortage of watts and wafers needed for data centers, (2) TSMC's limited willingness to expand capacity, and (3) deep scars from the 2008 financial crisis that keep a lid on tech valuations. These constraints may naturally limit overbuild.

Tech multiples compressed since 2025 start

Baker notes that tech multiples have actually compressed since the beginning of 2025, and tech now trades at a discount to staples—a rare occurrence. This suggests the market is not pricing in excessive AI optimism.

AI, Creativity, and the Future of Work

Human creativity may not be irreplaceable

Baker acknowledges Elon's concept that humans may be the biological bootloader for digital superintelligence. He suggests that human creativity may not be irreplaceable, and that finding new sources of meaning beyond work will be important as AI advances.

AI will transform video game development economics

AI world models will reduce AAA game development costs by 90%, from $100-300 million to much lower levels. This will increase competition among game developers but benefit video game platforms through an explosion of content. However, rendering will remain local on GPUs for the foreseeable future due to computational costs.

Neuralink as human-AI interface solution

Baker sees Neuralink as addressing humanity's IO problem—the limitation of speech, writing, and typing. By enabling faster information input and output, Neuralink could help biological intelligences coexist with digital intelligences and maintain value in an AI-dominated world.

Human brains are energy-efficient compared to AI

The human brain operates on 20-30 watts while training AI models requires hundreds of megawatts and inference requires far more than 20-30 watts. In a watt and energy-constrained world, Baker is optimistic that human intelligence will retain value for a long time.

Investment Philosophy and Mindset

Find philosophy and process that fit your emotional makeup

Baker emphasizes that successful investing requires finding an investment philosophy and process that aligns with your unique emotional makeup so you can remain rational when wrong. Most investors start as value investors because they've read Graham and Buffett, but this may not suit everyone's temperament.

Being early is the same as being wrong

Baker quotes Fidelity PM George Vanderheiden: 'Being early is the same thing as being wrong.' If you're losing money, you're wrong—not the market, not other people. This mindset forces accountability and clear decision-making during drawdowns.

Either panic early or double down late

Baker's mentor Jennifer Yuric taught him that as an investor, you must either panic early or double down late—almost no one does both. Knowing yourself means understanding which type you are. Baker identifies as a 'double down late' person, which shapes how he navigates tough periods.

Hardship early in career is advantageous

Baker argues that going through a major drawdown or failure early (he experienced this at age 25 on pharma) is actually advantageous because you learn resilience when the stakes are lower. The earlier you encounter hardship, the better you navigate it later when consequences are higher.

Decision quality improves after multiple drawdowns

Baker's goal is to reach a point where decision quality actually improves during drawdowns because he has data from past difficult periods. Instead of being gun-shy, he becomes more confident because he has proven he can make high-quality decisions under pressure.

Advice for Young Investors and Professionals

Be kind and be scrappy

Baker's core advice is to be both kind and scrappy—most people excel at one but not both. Kindness creates karma and attracts like-minded people who cooperate. Being scrappy means persistence and hustle. Together, they create a network of mutual support.

Your reputation is made in your early 20s

Baker cites Steve Schwarzman's biography: people remember how you behaved in your early career, not just your performance. Relationships formed early can resurface 20 years later when expertise is needed. This makes treating peers well a strategic investment.

Persistence and value delivery unlock mentorship

To get mentored by someone you don't know, Baker recommends: (1) be persistent and don't give up quickly, and (2) deliver value before asking for value. Try to know something the mentor doesn't, or offer insights they can use.

Only be an investor if you love the game

Baker emphasizes that public markets investing requires loving the game because you will go through really tough periods. If you can't find emotional and psychological centeredness during your first major drawdown, consider private equity or other paths with less volatility.

Seek hardship early in your career

Baker references the Sun Eater sci-fi series mantra: 'Seek hardship.' No matter how good you are, you will encounter it. The decisions you make during those times will define your career. Better to encounter it early when stakes are lower.

Elon and Mission-Driven Companies

Mission-oriented companies attract exceptional talent

Baker argues that Elon's success stems partly from creating mission-oriented companies. Talented engineers can choose between working on incremental ad clicks for Google/Meta or decarbonizing the world (Tesla), making humanity multi-planetary (SpaceX), or enabling free speech (X). Authentic missions attract exceptional people.

Tesla accelerated EV adoption by 20-30 years

Baker credits Tesla and Elon with accelerating EV adoption by 20-30 years and doing more for decarbonization than all environmental activists combined. This acceleration has had massive positive impact on the environment and the world.

Historians will view this as the age of Elon

Baker believes historians will look back on this era as the age of Elon because of his continued work on multiple world-changing missions (Tesla, SpaceX, Neuralink, X) while most centibillionaires have stopped working. Without Elon's continued effort, there would be no Starlink and lower odds of seeing a blue sunset on Mars.

Notable quotes

Being early is the same thing as being wrong. If you're losing money, you're wrong. — Gavin Baker
You either have to panic early or double down late. And essentially, no one does both. — Jennifer Yuric (via Gavin Baker)
Seek hardship. The decisions you make during those times will define your career. — Gavin Baker

Action items

  • Develop a touchstone (book, movie, or practice) to return to during market drawdowns and tough periods
  • Create a hobby or measurable goal (running, weightlifting, gaming) that lets you improve a number during difficult performance periods
  • Document your investment philosophy and process in writing, aligned with your emotional makeup, before you need it
  • Review past journal entries and decisions from previous drawdowns to build confidence in your decision-making under pressure
  • Practice being kind and scrappy simultaneously: build a network through genuine relationships while maintaining persistence in pursuing opportunities
  • When seeking mentorship, deliver value first before asking for it—share insights or knowledge the mentor doesn't have
  • Read 'A Wizard of Earthsea' and 'The Perfect Storm' as touchstones for navigating arrogance and perseverance through hardship

More like this