We're at the beginning of a landmark moment for Longevity. The companies built in the next 10 years will power a radical transition from sickcare to true healthcare. Here's a short documentary w/@omri_drory & friends about the future of #Longevity: youtu.be/sGoJV933sL4?si=s5y-…

Something that didn't make it into our "Screenless Startup" piece: The fascinating history of "invisible technology" in Silicon Valley. From proto-iPads to AI agents -- an active subculture has always believed technology should deliver value, then get out of the way.

Software used to fight for your attention. Every app. Every notification. Every screen. That was the product. Now the best AI products don’t ask for your attention — they remove the need for it. Watch how we're entering the era of the screenless startup 👇

Favorite @NFX essay in a while!

We are now deploying intelligence at scale. This changes the software game entirely. It's not a bigger version of SaaS -- it's a different game with stakes 100x larger.

The longevity breakthrough isn't supplements or biohacking. It's about technologies that can renew organs and push human lifespan to 160 . These breakthroughs exist today. @omri_drory reveals the key drivers:

Love this piece so much:

We're now investing out of NFX Fund IV. Like you, we're operators and entrepreneurs. And we see what you're seeing: The world of technology is at a turning point.

Latest video from @NFX on fundraising seasons for all you founders in the process:

Latest video from @NFX - send to your founder friends:

Top founders are re-thinking their go-to-market in the age of AI. That's why we released our upgraded go-to-market playbook. We got great feedback on it. So we reimagined it into a 5 minute video masterclass:

The old marketing playbook feels dated. Channels are saturated, expensive, inauthentic... But we're seeing startups scale faster than ever with leaner teams, and no traditional sales. There’s a new GTM playbook. x.com/NFX/status/19128708456…

Startups don’t need to do everything. They need to manage everything AI can do.

There are now 32,000 active VCs, up from just 150 thirty years ago. More competition, more capital, and better terms for founders. In our latest NFX Video Essay with @jamescurrier, we break down "Venture Capital 3.0"—where the industry is headed and what it means for you:

AI founders: A great product isn’t enough—you need stackable business models that compound value over time. Here’s @peteflint on building a defensible AI business for long-term success:

When we first published @jamescurrier's essay on consumer tech, the response from our network was incredible. Our new video essay dives deeper. We're breaking down the past, present, and future of consumer with more context, more clarity, and more actionable takeaways:

The VC ecosystem is changing. AI, mega-funds, more "surface area" for investing...the implications are huge for founders. It's mostly good news, but you have to see the whole picture. Welcome to "Venture Capital 3.0" by @jamescurrier nfx.com/post/venture-capital…

You can copy code, but you can’t download a rocket. stokespace.com/stoke-space-a…

C'mon people, are we really surprised by @deepseek_ai? Did you really not see it coming? From our @nfx blog in 2022: "The [core AI models] are going to be open-source and easy to use. Models is where the generative AI revolution started – but it’s not where it will end. It’s not going to take long, maybe 3 years, for models to approximate each other and reach human limits. The ability for these models to differentiate themselves stops at those limits. Data network effects asymptote over time. An AI model that is 5% or even 20% percent better than the competition is a pretty slim defensibility. The AI models will be switched out at will. Which of course ... commodify them.” There will be several more 100X reductions in speed and cost in the next two years, and eventually it will run on CPU's and we won't need GPU's, and it will run on your smartphone and laptop without heating it up. Note that @Meta pulled the same trick as DeepSeek but not as well (yet) with Llama which is to say "Fuck, I lost, so I'll blow the whole market up by doing opensource 6 months after you. If I can't win, I'm going to keep you from winning." Many players have it in their strategic insterest to open source AI capability. Meta is one. China is a second. There are many more. It's inevitable. Hasn't this been obvious from the beginning? The defensibility of these models companies is not to have the proprietary model that is better than another. That advantage asymptotes within a few years. They have a short window to create network effects on top of those models with workflow software, or security layers, or marketplaces, or 10 other methods. But that window closes. DeepSeek or the next DeepSeek, whoever it is, signals the closing of their window. These big dick swinging $500B projects, "I'm good for my $80B," or whatever, just got caught up in the competition with each other, ignoring the reality of the technology, misallocating VC money, shareholders money, and now trying to get taxpayer money. Once again, faster and smarter beats bigger. And look, if DeepSeek turns out to be not as great as we think now, don't worry, the next DeepSeek will be. And the next.

NFX has followed the AI agent economy from the start. Today, Partner Sarai Bronfeld discusses long term strategy insight from Israel: "The brain isn't enough anymore, you need action. Once AI can act independently, we are never going back." nfx.com/post/ai-agent-revolu…