How David Sachs thinks about AI & Crypto
2024 Edition from All-in Podcast.
Note: We are about to publish Q4, GenAI Safety and Security. And 2024 Year edition. If you have any comment to improve content, we’re happy to hear. - Emma
Only one month before the Trump Presidency, we’ve already felt a heat of the uncertainty with US Government Almost Shutdown event last Friday.
Looking back, it’s very helpful to know more about his head advisor of Science and Technology, David Sacks. He was well-known for the Elon Musk previous startup, Paypal. Later, he became an investor of an early stage venture, also started a few more startups and co-hosted a popular podcast called “All-in”.
This post is review based on his thought shared with an weekly audience in 2024. Hope it will give you the insight for what’s coming up in 2025.
*Discloser: Writer is a long time passionate listener of the podcast since episode ONE.
Profile
David Sachs, a venture capitalist at Craft Ventures and regular on the "All-In Podcast," is a multifaceted figure. Known as "Rainman," "the architect," and even a "Sith Lord" (The enemy of Jedi, an antagonist in a series Blockbuster Star wars) on the podcast, he's known for his contrarian views and "red meat" takes, often debating on tech, politics, and economics. Sachs co-founded PayPal with Elon Musk.
Described as a "moderate independent thinker," he's contributed to Fox News, is scheduled to speak at the RNC, and has been called "MAGA," yet also advocates for diplomacy in the Ukraine-Russia war. He's been jokingly called a "war correspondent" despite no military experience, and also labeled a "retribution officer."
While praised for his sharp, decisive management style, he's also been controversially called the "most evil person in Silicon Valley" by Paul Graham (Y Combinator co-founder) and accused of mistreating a founder. Sachs enjoys chess, Blackberries, and white wine, often seen swirling his glass on the podcast. He's known for his "tight haircut," the phrase "let your winners ride" (which he's "open-sourced" to fans), and even appeared as an AI version of himself on one episode.
"let your winners ride"
David Sachs’ s main points about AI & Crypto
Artificial Intelligence
Rapid Advancement: AI is expected to advance exponentially, with ongoing innovations reaching consumers, though predicting specific breakthroughs is difficult.
AI Models: Training AI models demands significant computing power, while inference requires speed and cost-efficiency.
Open Source vs. Closed Source: Open-source models may eventually reduce the economic value of AI models, even as their utility increases significantly. The release of open-source models comparable to leading proprietary models supports this trend.
Business Applications: AI holds the potential to create new markets, transform industries, and function as virtual senior employees or personal assistants, providing contextual support and task management.
Productivity: AI is anticipated to be a major driver of productivity gains.
Regulation: Government overregulation of AI is a concern, with a preference for legal precedents regarding fair use to be established before legislation.
Safety: Prioritizing AI safety over development speed might disadvantage companies in the race towards artificial general intelligence.
AI Agents: AI agents capable of decomposing objectives into tasks will be crucial for businesses, with significant potential for software-as-a-service offerings.
Mobile and Voice Assistants: Integrating large language models into existing voice assistants could significantly enhance their capabilities. There is also significant untapped potential for AI within the mobile space beyond chatbot apps and voice assistants.
AI and Existing Systems: Despite AI advancements, established systems of record will remain essential for businesses.
Cryptocurrency
Bitcoin: Bitcoin is gaining acceptance within traditional finance. 2024 is viewed as a pivotal year for Bitcoin, potentially due to ETF approvals. The impact of Bitcoin halving events on its price is acknowledged.
General
Data privacy, especially in encrypted messaging, is recognized as important.
AI Discussions
Open vs. Closed AI Models: Open-source AI models are championed, with the belief that their increasing utility will eventually diminish the economic value of closed models. The release of models like Llama 3 supports this perspective, contrasting with the view that proprietary models hold long-term economic advantages.
AI Development Pace: The exponential advancement of AI is expected to continue, despite the difficulty of predicting specific breakthroughs. Prioritizing safety over speed in AI development carries the risk of hindering progress in the race towards Artificial General Intelligence (AGI).
AI Regulation: Government regulation in AI is viewed with skepticism due to its potential to stifle innovation. Legal processes, like determining fair use, should precede regulation. State-level AI regulation is also seen as problematic.
AI's Business Impact: AI is transforming businesses, creating new markets and optimizing existing processes. AI agents capable of task breakdown are crucial. AI is projected to be a significant productivity driver. A single dominant AI platform is unlikely; smaller, interconnected models are anticipated.
AI Training vs. Inference: The distinct compute challenges of AI training and inference are emphasized. Inference, focused on speed and low cost, is poised to become a key development area.
Big Tech's Role: The assumption that only large tech companies can succeed in AI is questioned. Open-source approaches offer competitive advantages. Large tech companies, engaged in an "arms race," may be overspending on AI infrastructure.
AI and Mobile: Significant mobile opportunities exist beyond chatbot apps and voice assistants. AI assistants with environmental awareness are envisioned.
Siri's Future: Apple's major AI achievement will be integrating a large language model into Siri. This could divert search queries from other platforms.
AI and Systems of Record: Systems of record will remain essential for businesses.
Crypto Discussions
Bitcoin Adoption: Bitcoin's increasing mainstream acceptance and integration into traditional finance are acknowledged. 2024 is a pivotal year for Bitcoin due to potential ETF approvals. Bitcoin halving events and their impact on price are also followed.
Privacy: Data privacy, especially in encrypted messaging, is considered important.
Crypto Regulation: Excessive crypto regulation is criticized for its potential to stifle innovation. A reasonable regulatory framework is necessary.
"Sophisticated Investor" Tests: Implementing a "sophisticated investor" test could enable individuals to invest in crypto without risking significant financial loss.
Startup Coins: Creating a "startup coin" to allow public investment in venture capital portfolios is seen as a promising concept.
Recurring Themes
Innovation vs. Regulation: Regulation should not impede innovation, particularly in emerging fields like AI and crypto.
Open vs. Closed Ecosystems: Open-source models are favored over closed, proprietary systems.
Progress and Safety: Balancing the need for safety in AI with the importance of continued progress is key.
Tech's Societal Impact: The transformative effects of AI and crypto on society and the economy are central considerations.
Best Quotes about Gen AI Security and Safety
"if the internet had been regulated...in 1995 or 1997 it never would have blossomed the way that it did”
“when you have agents...you don't need to spend tens or hundreds of millions of dollars to wrap your revenue in something that says it's a system of record"
"you still need databases, you still need Enterprise security...you still need compliance...and Enterprise customers don't want to DIY it"
"I think maybe the best you can do is is Advance AI to be truthful"
"I would spend a hundred billion dollar a year licensing data and then I would present the truth"
"let your winners ride"