Survival skills for modern founders

Does money make us less biased?

Betting markets are the new news.

How Glean hacked enterprise search.

Why trading systems are rigged

Are you losing the AI race?

Why AI agents fail at long tasks.

Why building is worth the barrier.

How AI noise killed mass recruitment.

Why the best engineers waste their lives

• Young workers in AI-exposed roles are currently experiencing a 16% decline in employment growth compared to their less exposed peers.
• Contrary to some fears, there is currently no evidence that interest rate sensitivity or tech-sector over-hiring fully accounts for these downward employment trends.
• Strategic thinking, high-level social interaction, and complex guidance are identified as the core human skills that will define value in an automated future.
• AI has the massive potential to foster a 'career lattice,' allowing workers to pivot between professions more fluidly as industry demands evolve.

• China maintains a significant advantage in industrial infrastructure, manufacturing speed, and energy capacity, placing them far ahead of the U.S. in physical production capabilities.

• While the U.S. leads in core AI model innovation, this advantage may only be moderate, as the Chinese ecosystem is catching up rapidly through efficient open-source and local model development.

• Both nations are currently hindering their own progress through self-inflicted strategic errors, such as restrictive social engineering in China and protectionist, exclusionary policies in the United States.

• Competitive success belongs to the nation that can move away from excessive central interference and focus on fostering, rather than hindering, its domestic economic and innovative potential.

• Shift from treating AI as an intern toward viewing it as a specialized assistant that handles strategy and complex tasks.
• Future professional success depends on developing 'Super Individual Contributor' skills that combine strategic direction with hands-on output.
• Start by building simple, single-purpose AI agents rather than complex systems, and iterate based on real-world effectiveness.
• Career stability now comes from continuous learning and skill growth rather than staying within one large, static organization.

• Shift your perspective from using AI as a mere technical tool to treating it as a digital coworker integrated into your workflow.

• Mastering the art of collaboration with AI is a social skill that requires ongoing practice, repetition, and an understanding of its unique capabilities and constraints.

• Voice data represents a vast, untapped frontier of human knowledge that remains largely uncaptured.

• Startups should build proprietary deep technology rather than relying on third-party APIs to ensure long-term differentiation and lower costs.

• Cultural resistance to recording meetings is a hurdle that eventually gives way to productivity gains as adopters see clear value.

• Voice is poised to become the dominant interface for business intelligence, eventually reducing the need for keyboard-based writing.

• Shift your mindset from immediate task automation to long-term R&D investments that build future capabilities.

• Accept that early experimentation with experimental AI workflows may temporarily decrease individual productivity.

• Relying on AI tools for transactional tasks can lead to skill atrophy, as demonstrated by a 17% performance drop in follow-up assessments when users were deprived of AI assistance.

• The quality of learning depends on user intent; those who engage with AI inquiry probes rather than relying on it for completion outperform those who skip the cognitive struggle.

• Emerging market users who engage with AI as their primary technology often grasp its full potential more effectively than those tethered to outdated assistant-based models.

• Adopting an AI-native perspective requires abandoning legacy definitions to focus strictly on what current and future models are truly capable of achieving.

• Shift your mindset from treating AI as a simple assistant to viewing it as a sophisticated, evolving collaborator that requires social skills to manage.
• Prioritize providing deep context and complex, open-ended problems rather than simple, transactional queries to maximize model performance.
• View the use of AI not just for task completion, but as a critical laboratory for 'skill-building' and personal growth, balancing AI use with manual practice to avoid cognitive atrophy.
• Develop the ability to build and deploy custom AI agents, which should be viewed as an essential professional competency similar to spreadsheet proficiency.

• Success in AI comes from solving deep, specific operational problems rather than relying on automated surface-level shortcuts.
• Founders should focus on 'falsifying' their startup ideas early to save time, rather than seeking validation for unproven concepts.
• Developing a deep professional understanding of a domain allows AI to act as a powerful accelerator, whereas using AI to bypass fundamental learning creates fragile tools.