Andreessen Horowitz

Diode Computers, Inc. is using AI to design circuit boards, simplifying messy, slow workflows. We’re excited to invest in Diode to scale manufacturing of hardtech in the US.   Circuit boards are in every smartphone, medical device, and fighter jet. Better circuit design means more of the tech we rely on. Full blog post here: https://lnkd.in/g-QtZsgb None of the above should be taken as investment advice or an advertisement for investment services; please see https://lnkd.in/gjmYWkHi for more information. For a full list of portfolio companies, visit https://lnkd.in/g7hjSuHb.

It’s here! I’m thrilled to announce the Navan Business Travel Index (BTI for short)! Investors, CFOs, finance leaders and data nerds (like me) need to check out the full report here: navan.com/bti Drop a comment about which findings you think are most interesting or surprising (hint: I like the compare versus TSA data)! 📊

Founders keep asking: What does real breakout velocity look like in AI? This issue breaks down the patterns we’re seeing across product, growth, GTM, and defensibility from leading enterprise AI companies. It features takeaways from our latest article, From Demos to Deals, plus conversations with leaders from Box, dbt Labs, Metronome, Labelbox, Prepared, Vanta, and Arcjet and deep dives on GTM, infra, data, and more. Some takeaways: - “Flashy demos are easy. Substantive products are hard.” - 10X is the new 3x: “The fastest growing AI software companies today are growing more than 10x year over year.” - “AI software often sells the work output itself, instead of selling software to help the people do the work.” - “There’s an incredible amount of value to being the first to establish yourself as that provider.” - “AI itself is not a moat: it is a way to deliver value to customers.” - “AI is changing the value prop of software.” AI shifts the buyer calculus from access to output and business models have to follow. (Scott Woody) - “Analytics always expands to fill the available budget.” Cheaper compute and better tools are creating entirely new demand for data-driven decision-making. (Tristan Handy) - “The world is shifting from building AI models to renting AI intelligence.” Teams don’t want to train models, they want modular, enterprise-grade AI they can plug into. (Manu Sharma) Check it out and subscribe:

Huge day for Ash by Slingshot AI. Ash is the first foundation model for real therapy designed to do what most AI won’t: challenge you to grow. Created to help a billion people access meaningful mental health support. Congrats to Daniel Reid Cahn, Neil Parikh, and the team.

Our infra team breaks down the concept of “context engineering”: Martin Casado: “If you are going to call a model, you have to know what to put in the context… At some point you’re probably going to use traditional computer science." Jennifer Li: “What’s the new form factor of infra that needs to become part of this context engineering?... How do you have agents use tools, or infrastructure that will provide discovery and guarantees of observability of these systems as well?” “New infrastructure pieces create new patterns and methods of software and building systems. This is a great example of it emerging before our eyes.”

Welcome to the era of self-eating software. In the old world, software disrupted taxis, sales, and back office. In the new world, AI is disrupting software itself. From our GP Martin Casado: “The profession that I’ve dedicated my entire life to is being disrupted, and it’s very exciting because it’s kind of eating itself.”

The Medicaid world needs Fortuna Health more than ever: because of the relentless pace of federal and state Medicaid policy changes over the last few years, Medicaid eligibility is getting way more complex. A lot of the chatter has been focused on reduced eligibility, but there's an equally important challenge associated with those who are eligible but lose out on coverage because of administrative process complexity - in the U.S., losing health coverage can boil down to something as simple (and unfortunate) as missing a letter in the mail. Fortuna is fixing this. The company launched in the throes of redeterminations to make Medicaid eligibility determination and enrollment clear, navigable, and reliable. We couldn't be more excited to lead Fortuna's Series A (after also leading the Seed) and continue backing this incredible, run-through-walls team - Nikita Singareddy, Cydney Kim, and Benjamin Wesner - as they build clearinghouse infrastructure and consumer co-pilots for Medicaid – and beyond. More here: https://lnkd.in/gBjixSux Daisy Wolf, Eva Steinman, Jane Rhee Andreessen Horowitz a16z Bio + Health

President Nixon tried to scale nuclear energy in America, but failed. From a16z co-founder Marc Andreessen: “He saw the energy crisis coming.” “ He said it's a national imperative to build 1,000 new nuclear power plants.” “ The bad news is, he created the EPA and the Nuclear Regulatory Commission.” “[This] then prevented Project Independence from happening." (From National Economic Forum with Joe Lonsdale)

“AI is changing the value prop of software.” We spoke with Metronome CEO Scott Woody about why AI requires a new business model. “In the cloud era…the value of the software scales as your org scales.” “In AI, the software is now doing work on your behalf…so the core value of the software is not who in my org has access to the data, it's what can this software do for me? When the value shifts from access to output, the business model has to follow.

US per capita energy usage peaked in 1973. China's has increased 9x since then. We forgot how to build mega projects. Our grid has ossified. Our partner Ryan M. explains: