The Spotlight by Founders Capital - Edition 8

What We’re Reading

The VC industry needs a geopolitical reboot - Financial Times

Venture capital is at a turning point - slowed deal flow, capital-heavy AI, and rising geopolitical pressure mean the old Silicon Valley playbook no longer works. In this punchy FT op-ed, John Thornhill argues that if Western VC wants to stay competitive, it must evolve into a more strategic, government-aligned force, much like the models emerging in North-East Asia.

The controversial collapse of a £1 bn British AI dream - The Telegraph

The Telegraph tells the story of Builder AI’s sudden collapse, once a $1.5bn poster child for UK tech, now in administration. Backed by Microsoft and led by the flamboyant Sachin Dev Duggal, the company’s fall exposes the gap between image and reality in high-growth startups, from inflated sales claims to mounting governance concerns.

Private Markets Are Crowding Out Public Markets. Where’s the SEC? - Barrons

A critical take on how private markets are eclipsing public markets: there are now ~4,000 public U.S. companies left, while private equity, credit, and VC dominate headlines. With retail biting into private deals, the article argues the SEC’s silence on oversight could spell trouble.

Must-Listen Podcasts

Bill Gurley, The Gift and The Curse of Staying Private - Invest Like the Best

If you want a clear-eyed take on today’s venture landscape, Bill Gurley (legendary Benchmark Capital VC) delivers. He breaks down why returns are tightening, why companies are staying private longer, and how misaligned incentives between LPs, GPs, and founders are warping the game.

Varun Mohan, CEO and Co-Founder of Windsurf - 20VC

Varun Mohan unpacks why moats are fragile in an AI-driven world, why founders must be willing to kill their own ideas, and how LLM commoditisation is reshaping the software stack. A sharp take on building fast and staying relevant in the age of AI-native tools - from one of the hottest AI founders right now.

Founders Capital Perspective

If you’re enjoying this newsletter and want to stay on the pulse, make sure to read our weekly private market trends newsletter - The Dealflow by Founders Capital. Last week we discussed Wise ditching the LSE, and, as always, shared our insights and perspective.

Research & Resources

The Illusion of Thinking: Understanding the Strengths and Limitations of Reasoning Models via the Lens of Problem Complexity - Apple

In a widely shared new paper, Apple researchers tested the limits of large reasoning models (LRMs) using logic puzzles such as Tower of Hanoi, Checker Jumping, River Crossing, and Blocks World - and found that as problem complexity increased, model performance collapsed. While LRMs handled simpler tasks well, they often failed outright on more complex ones, even when given enough compute and the correct algorithm. Strikingly, as complexity rose, models didn’t ramp up effort, they often exerted less, pointing to deep flaws in how current systems allocate reasoning and manage multi-step logic.

The paper has (shockingly) sparked debate. Critics argue Apple’s tasks are too synthetic and don’t reflect how models perform in real-world settings. Others see it as a strategic move from a company still playing catch-up in generative AI, especially as Apple is rumoured to be integrating third-party models into its ecosystem. Either way, the research casts real doubt on how much progress we’ve actually made toward machines that can reason in a meaningful way.

Last week we asked you: What’s the real bottleneck in VC right now?

The most popular response was: “(A) Lack of liquidity - LPs just don’t have the capital”

Beyond the Boardroom

Adam Grant Says It's Good to Rethink About Things  – Dartmouth

Adam Grant unpacks the “I’m-not-biased” bias, showing how our belief in our own objectivity can blind us to better ideas. The core takeaway? Create space for dissent and curiosity, because good thinking starts with unlearning.

Do Things That Don’t Scale – Paul Graham

Do Things That Don’t Scale is a must-read for anyone building something from scratch. He argues that the early stages of a startup are all about doing the gritty, unscalable things, like personally onboarding users or hand-crafting experiences, to truly learn what works.

Like what you see? Think we missed anything? Let us know and we will check it out ahead of the next edition.