This episode discusses how to spot future tech trends. Chris Dixon (a16z partner) says to watch what smart engineers do on weekends — they're now studying crypto while working at Google/Facebook. He favors open networks (rules don't change) over closed platforms like App Store that kill innovation. Key holdings: Coinbase (early investment, crypto wallet/exchange), Ethereum (blockchain for smart contracts, currently 10 transactions/second but improving), Bitcoin (created in 2009, split over scaling debate).
Chris Dixon (Partner at Andreessen Horowitz) discussed future technology trends on the Invest Like the Best podcast. His core thesis is that observing what smart people do on weekends can predict the technological direction years ahead. He focused on cryptocurrencies, autonomous driving, and drones,
Chris Dixon (Partner at Andreessen Horowitz) is one of the most insightful thinkers in technology investing, having shifted from philosophy to tech investing. The core themes of this episode: methodology for judging technology trends, the game of network design (open vs. closed), and cryptocurrency as a new paradigm to break the monopoly of tech giants. The most impactful takeaway: "Observing what smart people do on weekends can predict the direction of technology years in advance" — Dixon believes that the areas engineers dedicate their free time to often represent major opportunities over the next 5–10 years.
Chris Dixon proposes a practical rule for identifying technological trends: observe what smart people do on weekends.
Dixon recalls his experience during the Web 2.0 era (2003-2009): during the day, people worked at enterprise software companies, while at night and on weekends, they discussed "toys" like Flickr, YouTube, and Friendster. These "toys" ultimately became world-changing platforms. He observes the same pattern recurring in the cryptocurrency space—engineers at Google and Facebook optimize ads during the day and study Ethereum at night.
> "The hobbyists at nights and weekends, those people are thinking on like a five to ten year horizon, right, by definition."
This means: hobbyists at night and on weekends think about the long-term future of 5-10 years, while their bosses focus only on quarterly goals of 1-2 years.
Dixon argues that engineers vote with their time—if a large number of smart people dedicate their spare time to a particular field, that is the strongest forward-looking signal.
Dixon argues that the key to the internet's success lies in its open protocol architecture. Basic protocols like TCP/IP are maintained by non-profit organizations, and developers know the rules will not suddenly change. This gave rise to giants like Google and Facebook—Larry Page and Mark Zuckerberg could start their companies in dorm rooms because they were confident the internet would not suddenly "change the rules and charge fees."
> "The internet was a set of decentralized protocols... it was like the highway system. If you built this awesome trucking company on the highway system, the highway system wasn't going to come along and take it from you."
Meaning: The internet is a set of decentralized protocols, much like a highway system—if you built a great trucking company on the highway, the highway would not come along and take your company away.
Dixon points out that the situation today is vastly different:
Core Concern: Can you still start a company in a garage today that challenges Google or Facebook? Dixon believes the answer is increasingly uncertain.
| Feature | Closed Platforms (Google/Facebook) | Open Protocols (Internet/Crypto) |
|---|---|---|
| Performance | High (centralized optimization) | Low (Ethereum supports only 10 tx/s) |
| Innovation Incentive | Developers fear rule changes | Developers are confident rules won't change |
| Developer Ecosystem | Limited by platform policies | 20 million developers worldwide can participate |
| Value Distribution | Platform captures most value | Protocol layer/users share value |
Dixon outlines the development trajectory of cryptocurrencies:
Dixon argues that the most important feature of cryptocurrencies is not "censorship resistance", but "rule immutability"—users and developers can be confident that the platforms they build on will not suddenly change the rules to strip them of their value.
This aligns with the advantages of early internet open protocols. However, Dixon also candidly points out current limitations:
Dixon observes an interesting shift: traditional VCs are being "marginalized" in the cryptocurrency space. Projects like Ethereum and Augur raise funds directly from the community through ICOs, bypassing the need for VCs. Andreessen Horowitz's response strategy includes:
1. Investing in cryptocurrency hedge funds (e.g., Polychain)
2. Directly purchasing protocol tokens (rather than company equity)
3. Shifting evaluation criteria from "team + market" to "whitepaper + code + community"
Dixon believes the three most important technological directions for the next decade are:
| Trend | Current Stage | Key Signals |
|---|---|---|
| Artificial Intelligence | Breakthrough phase (ImageNet error rate already below human level) | Applications of deep learning in healthcare, autonomous driving, etc. |
| Cryptocurrency | Infrastructure phase (similar to the internet in the 1990s) | Performance improvements (from 10 transactions/second to scalable) |
| New Computing Platforms | Early stage (VR/AR/autonomous driving) | Declining hardware costs, improving user experience |
Dixon acknowledges that by the time a trend becomes consensus, the investment opportunity may have already passed. He uses VR as an example:
Key insight: Cryptocurrency is still in a "contrarian" state—Wall Street titans (such as Jamie Dimon) publicly mock it, and major tech companies have yet to enter aggressively. This is precisely why Dixon finds it attractive.
Dixon points out that 20 years ago, people expected the internet to fundamentally transform corporate organizational forms (an application of Coase's theorem), but the reality is:
Dixon argues that cryptocurrencies showcase a new form of organizational structure:
> "It's this thing that's whatever the current market cap is... and there's no organization behind it."
In other words: It is an entity with a massive market capitalization, yet no organization behind it.
Dixon hopes this represents the true realization of the internet's promise of "loosely connected small fragments"—where future work may be coordinated through market mechanisms and network protocols, rather than traditional corporate hierarchies.
Dixon cites data from the "Our World in Data" website to counter automation fears:
| Indicator | 200 Years Ago | Now |
|---|---|---|
| Global poverty rate | Extremely high | Significantly reduced |
| Infant mortality rate | Extremely high | Significantly reduced |
| GDP per capita | Extremely low | Significantly increased |
Dixon argues that the view that automation will eliminate all jobs is a "lump of labor fallacy"—the assumption that demand is fixed. In reality:
> "Human needs and wants are unbounded."
Meaning: Human needs and desires are unlimited.
Historical pattern: New technologies make it easy to see jobs that disappear (e.g., printing press workers), but difficult to predict newly created jobs (e.g., Uber drivers, social media managers, mobile app designers). No one could have predicted these occupations 20 years ago.
Dixon does not offer specific career advice but implies:
| Position | Guest Stance | Key Data |
|---|---|---|
| Coinbase | Bullish (early investment) | Invested in 2013; cryptocurrency wallet/exchange |
| Ethereum | Bullish (infrastructure layer) | Supports 10 transactions/second; ERC-20 standard sparked ICO boom |
| Bitcoin | Neutral (technical discussion) | Launched in 2009; Block Size debate led to a split |
| Oculus | Bullish (early investment) | Invested in 2013; later acquired by Facebook |
| Polychain | Bullish (indirect investment) | Cryptocurrency hedge fund; co-led by a16z |
| Positive case | 50-person team; acquired for $26 billion | |
| Positive case | Smaller team; acquired at high valuation | |
| TBH | Risk warning | Acquired by Facebook for $100 million; Dixon views this as a case of stifled innovation |
1. The "Weekend Test" Principle (Chris Dixon): Observing what smart people do in their spare time can predict technological trends 5-10 years ahead. Cryptocurrency currently fits this pattern—Google/Facebook engineers optimize ads during the day and study Ethereum at night.
2. The Fundamental Difference Between Open Protocols and Closed Platforms (Chris Dixon): Open protocols are like highways—you can build a trucking company, and the highway won't compete with you. Closed platforms are like private toll roads—they can change the rules at any time and charge a 30% "toll."
3. The Core Value of Cryptocurrency Is Not Censorship Resistance, but Rule Immutability (Chris Dixon): Developers are willing to build on Ethereum because they are confident the rules won't suddenly change—this echoes the advantage of open protocols in the early internet.
4. The "Degradation" of the Internet (Chris Dixon): None of the top 20 apps in the App Store (excluding games) were created after 2012, indicating that the innovation pipeline has been clogged by closed platforms. Today, it is difficult to build a company from a garage that can challenge Google or Facebook.
5. Cryptocurrency Is in the "Infrastructure Phase" (Chris Dixon): Similar to the internet in the 1990s—poor performance (Ethereum processes only 10 transactions per second), high barriers to entry, but rapidly improving. There are 5-10 clear technical paths to solve the performance issues.
6. Investing Requires Being "Contrarian and Correct" (Chris Dixon): When a trend becomes consensus, the opportunity may have passed. Cryptocurrency is still in a contrarian state (Wall Street titans openly mock it), which is precisely its appeal.
7. The "Unorganized" Form of Organization (Chris Dixon): Bitcoin has a massive market cap but no company behind it—only a group of developers arguing on GitHub. This may be the true realization of the internet's promise of "loosely connected small fragments."
8. Automation Will Not Eliminate Jobs (Chris Dixon): Human needs are infinite. New technologies make it easy to see jobs that disappear (e.g., printing press workers), but hard to predict new jobs created (e.g., no one could have predicted Uber drivers or social media managers 20 years ago).