The hottest Tech Investing Substack posts right now

Paying 15× NAV for VCX just to get Anthropic exposure is an insanely rich valuation.
There are multiple far better options to gain exposure to Anthropic and OpenAI than overpaying for VCX.
If you insist on paying extreme multiples, at least structure the deal so you also get a well-known legacy business alongside the AI bet.

Investors are simultaneously dumping SaaS stocks and AI infrastructure stocks because they fear two opposing things at once: that AI will replace software businesses and that AI spending won’t deliver returns.
A recent leap in AI capabilities that lets models handle tasks like legal, finance, and marketing convinced traders that AI can move into the application layer, which sparked the selloff in software companies.
The market’s mixed selling is a rational response to deep uncertainty: if AI truly upends software then heavy infrastructure buildout is justified, but if it doesn’t then that spending looks wasteful, so investors hedge by selling different parts of the ecosystem.

A major AI data‑center expansion lost its anchor tenant after financing and changing customer needs, showing that big buildouts can stumble once the real math replaces slides.
Chipmakers and hyperscalers are stepping in to protect GPU demand—Nvidia put down a large deposit and helped recruit a tenant—so suppliers may finance infrastructure to safeguard sales.
That hiccup comes amid Iran tensions, private‑credit stress, and positive real rates, meaning a crack in the crowded AI capex trade could amplify market volatility.

Buying beaten-down public SaaS stocks right now is risky because industry-wide malaise can persist and you can get whipsawed trying to catch falling knives.
Expect more dispersion: the market will keep punishing losers while only labeling survivors as winners in hindsight, so the real edge is identifying which companies will survive in real time.
Many software firms won't die but will become low-growth 'zombies', so be selective and favor businesses that can genuinely transition to and benefit from AI, using a disciplined checklist to rank longs and shorts.

AI is basically digital automation that can massively scale the production of digital content and actions. More supply doesn’t guarantee value — demand, human preferences, and political or social feedback will determine the real economic outcomes.
Lasting business advantage comes from data moats, ecosystems, and distribution, not just big models or hardware. Open-source models, model compression, and competition can erode hardware/software moats and make many pure GPU bets risky.
The best hedge is non-financial: invest in human advantages like relationships, health, and skills while diversifying attention and capital across other macro risks. Build human-centered products and networks that complement AI instead of relying solely on AI hype.

Get a weekly roundup of the best Substack posts, by hacker news affinity:

Modern AI systems run on huge models trained with massive datasets and require enormous compute — specialized GPUs, large data centers, lots of energy, and a concentrated global chip supply chain.
The current AI boom resembles past tech bubbles because vast infrastructure and speculative valuations risk collapsing if those investments don’t translate into sustained cash flows or viable business models.
Evidence of AI’s productivity gains is mixed and uneven: some tasks see modest improvements, adoption has plateaued in places, and public, political, and regulatory resistance (especially to data centers) could limit broader economic impact.

The recent software sell-off is partly a market overreaction, not the death of mission-critical SaaS. Incumbent vendors that adopt AI can protect pricing power and improve free cash flow.
Companies with "artificial limiters" — non-code moats like network effects, regulatory barriers, and physical infrastructure — are best positioned to re-accelerate growth and can become multi-baggers if bought at the right price.
Venture investing is riskier now because public multiples are compressed and many startups are still effectively SaaS, so private-market entry prices may not be justified by exits, making public equities a clearer place to find mispricings.

Techno-optimism in the V.C. world can signal a crisis.
The V.C. industry's emphasis on techno-optimism may reflect distress.
The proliferation of techno-optimist views may demonstrate underlying challenges in the venture capital landscape.

India’s tech scene is following a path similar to China’s around 2010, which suggests a big multi-year opportunity as local companies scale and markets mature.
The idea that "software always wins" is overextended—software valuations and expectations are cooling, so investors should be more selective and update their outlooks.
A rapid, raw approach to sharing investment ideas helps surface connections between theses and exposes where real conviction (or doubt) lies.

The internet doesn't have feelings. Don't treat it like a person.
Handwriting is important for memory and cognition. Maybe we should write more.
Predicting the next big tech thing is challenging but possible with the right strategy.