Tomasz Tunguz
Showing 1–7 of 7 transcripts.
Peter Thiel and Softbank Sell NVIDIA - Why? & Why VC Will Hit $1TRN and The Opening of Retail
Peter Thiel, Tomasz Tunguz, Jason Lemkin, Rory O'Driscoll, Harry Stebbings
Cursor secured a $2.3 billion investment at a $30 billion valuation by promising 30–70% productivity gains for a developer market valued in the trillions, while relying on its lean 30-person structure to maintain high margins. The discussion contextualizes this deal within a volatile financial environment marked by AI supply constraints, macroeconomic stress signals, and a venture capital shift toward late-stage liquidity over traditional IPOs. Analysts project a future where three to four years of rapid disruption will consolidate the market, with Cursor competing against hyperscalers for dominance through deep workflow integration and proprietary model efficiency.
Inside Theory Ventures: Tomasz Tunguz’s $688M Thesis on Go-To-Market Disruption
Theory Ventures, led by Tomas Tungus, secured a $450 million Fund II in late 2024 to deploy $1–$25 million investments into early-stage companies exploiting AI-driven market shifts. Tungus forecasts that 2025 will trigger the first reliable mainstream integration of AI agents while identifying energy constraints and shifting search advertising models as critical industry dynamics. The firm's concentrated strategy targets three pillars—modern data infrastructure, labor-scarce workflow automation, and decentralized blockchain databases—supported by an in-house intelligence team managing research operations.
Sam Altman, Arthur Mensch and more discuss:Which Startups Are Threatened vs Enabled by OpenAI?|E1156
Sam Altman, Arthur Mensch, Brad Lightcap, Des Traynor, Tom Hulme, Tomasz Tunguz, Sarah Tavel, Harry Stebbings, Emad Mostaque, Tom Blomfield, Miles Grimshaw
Industry leaders including Meta, Mistral, and Google are driving a market consolidation where commoditized base foundation models shift competitive advantage toward deep workflow integration and personalized "thick wrappers." While cloud providers and incumbents leverage existing infrastructure to dominate the utility layer, startups face valuation risks and must differentiate through outcome-based business models rather than thin wrappers. The predicted outcome is a global oligopoly of five to six dominant model providers by 2027, forcing applications to evolve from seat-based licensing to selling full work products within specific verticals.
The Ultimate AI Roundtable: What Happens Now in AI, Why Google are Vulnerable | E1085
Des Traynor, Yann LeCun, Emad Mostaque, Jeff Seibert, Tomasz Tunguz, Douwe Kiela, Cris Valenzuela, Richard Socher, Harry Stebbings, Myles Grimshaw, Christian Lang
A diverse panel of AI industry leaders predicts a market consolidation around five to six dominant providers while debating the trajectory from proprietary scale to commoditized, open-source efficiency. Experts contrast conflicting views on model defensibility, with some arguing that data quality and iteration speed matter more than parameter counts, while others maintain that massive size remains essential for handling complex tasks. The discussion further outlines a structural shift in value accrual from cloud infrastructure to differentiated applications, forecasting a pivot in business models toward outcome-based pricing and predicting existential challenges for search incumbents like Google alongside opportunities for hardware-integrated strategies from Apple.
How I'm Investing My $230M Fund - AI Investor Tom Tunguz Breaks It Down
Tom Tunguz, Tomasz Tunguz, Harry Stebbings
A solo general partner is launching a $230 million fund targeting 12 to 15 portfolio companies with a high-conviction strategy where the top three holdings may represent up to 50% of the total assets. Leveraging a "decade of data" framework and Fermi estimation models from *Super Forecasters*, the GP focuses on realistic forward multiples near 5.5x rather than hyper-growth projections while reserving initial checks between $8 million and $12 million per company. This thesis-driven approach prioritizes six to twelve months of sector research and continuous probability monitoring to deepen convictions in data-centric ventures despite the inability to lead $50 million Series A rounds.
AI Investing 101: Tom Tunguz Breaks Down His Investment Thesis and How To Get Started
Tom Tunguz, Tomasz Tunguz, Harry Stebbings
The AI ecosystem is projected to bifurcate into a hybrid structure combining closed, capital-intensive foundational systems with a decentralized open-source layer, while consumer interfaces will increasingly mediate across multiple specialized models. Enterprise adoption will initially favor bundled, end-to-end solutions for Global 2000 firms lacking technical sophistication, driven by the need for secure data planes that remain isolated from cloud-hosted applications. As organizations mature, they are expected to transition toward modular stacks, fostering new business ventures focused on legal and operational risks such as copyright infringement and PII leakage.
Tomasz Tunguz: How I Raised $230M; ChatGPT vs. Google; How LLMs Work; Trump vs DeSantis | E1004
Tomasz Tunguz, Trump, DeSantis, Harry Stebbings
Tom, a solo general partner, successfully raised a $230 million fund to pursue thesis-driven investments in AI infrastructure, utilizing a concentrated portfolio strategy that targets 10% ownership in Series A companies. His investment thesis centers on the convergence of closed and open-source models, predicting that superior execution and unique data moats will allow startups to outmaneuver incumbents like Google and Adobe in the coming decade. The campaign achieved a single close by securing two Institutional LPAC members as anchors and leveraging data room analytics to convert 50% of limited partners from previously unknown relationships.