Matthew Lynley

Technical PMM @dMatrix_AI. I also run a blog called Supervised. Recovering data analyst, math @UNC. Prev: Sr. Data Analyst @Pluralsight, blogger @TechCrunch.

[Matthew Lynley @mattlynley]

Good morning! Today, I am launching a new newsletter covering the incredibly fast-moving AI industry: Supervised. https://t.co/C4UjOq46u7

[Dr. Jon Slotkin @slotkinjr]

The raw data CSVs can be found at the way bottom of this page

https://t.co/GTUFcHe1lA

[Dr. Jon Slotkin @slotkinjr]

As a neurosurgeon I care a lot about road safety.

By now you’ve probably seen @Waymo’s stunning safety results (like 91% fewer serious crashes). But they didn’t just publish data headlines. They released the raw CSV files and data dictionaries.
I did a much deeper analysis. A fascinating story emerges when you analyze how they’re achieving this.

This isn’t incremental improvement - it’s categorical. We’re looking at the potential elimination of traffic deaths as a leading cause of mortality.

The intersection breakthrough: Waymo has essentially solved intersection crashes, with 95% fewer injury incidents than human drivers in the same locations. That’s transforming the deadliest driving scenario.

The national math: If every US vehicle performed like Waymo, we’d prevent 33,000-39,000 deaths annually and save $0.9-1.25 trillion in societal costs. Even partial adoption at 27% would save ~10,000 lives per year. In terms of magnitude, this would be the equivalent of eliminating every pedestrian death nationally in a year.

The physics signature: Here’s what fascinates me: 47% of Waymo’s contacts involve less than 1 mph delta-V. They’re not just avoiding crashes; they’re converting unavoidable incidents into gentle bumps. It’s like having physics itself on your side.

We’re not talking about marginal safety gains. The data represents a fundamental shift from harm reduction to harm prevention.

The methodology matters: I used their dynamic geographic benchmarks (comparing like-for-like road conditions) and verified the findings hold across San Francisco, Phoenix, LA, and Austin. The safety advantage actually increases in more complex urban environments.

Link to raw data below….

Notes on my approach:

Analysis based on 96 million miles of Waymo Rider-Only (RO) data through June 2025, utilizing Waymo's dynamic geographic benchmarks to compare Waymo Driver performance against human drivers under similar road conditions and operational design domains.

The projections for national impact (deaths prevented, societal costs) involve several assumptions. Given Waymo's zero reported fatalities, the direct serious injury reductions were mapped to national fatality statistics using established NHTSA-derived ratios that correlate serious injury crash rates with fatality rates. This extrapolation assumes that Waymo's observed serious injury prevention capability would translate proportionally to fatality prevention. Societal cost savings are estimated by applying average per-fatality and per-injury economic costs (e.g., medical, lost productivity, quality of life) as published by NHTSA, scaling these national averages to the projected number of avoided fatalities and injuries based on Waymo's safety performance. These figures represent the potential annual impact if the Waymo Driver's safety profile were widely integrated into the national fleet.

@ethanteicher

[Matthew Lynley @mattlynley]

@tunguz @OfficialLoganK Needs an xgboost button next to the like button

[Matthew Lynley @mattlynley]

Good bot

[Matthew Lynley @mattlynley]

Excited to be starting at @dMatrix_AI as a Technical PMM! Efficient and high-performance inference will be a tremendously important part of the next phase of AI, from agents to ever-more powerful reasoning models. I'm thrilled to be working at the forefront of this next phase!

[Michelle Fang @michejafang]

Excited to share I have joined the eclectic team at @Lux_Capital.

Grateful for my time at Cerebras with Andrew and a team who showed me the vision, grit, and adaptability it takes to build a generational AI semiconductor company. The same curiosity that drew me from the Senate to Silicon Valley to understand the bare metal powering AI, has now led me to join Lux to partner, invest, and explore how AI is transforming the physical world.

Lux’s commitment to building technology at the frontier is special, and Josh and Peter have built a incredible team that I’m lucky to work with and learn from.

To founders and friends, especially in the robotics, manufacturing, and energy space, I’d love to meet you.

🚀@wolfejosh, Peter, @breeves08, @Farshchi, @deenashakir, @graceisford, @velvetatom, @lanjiang653, Shaq, @davidkmyang

[Matthew Lynley @mattlynley]

Full thing here (and why I had to end working full time on the newsletter) https://t.co/U56S4WFk0h

[Matthew Lynley @mattlynley]

I’m back to (recreationally) writing while figuring out what’s next. First thing looking what’s happening with the modern data (and MLOps) stack. With the flurry of M&A (and M&A talks), the long-coming and frequent joke that the MDS consolidation may have finally arrived!

[Matthew Lynley @mattlynley]

Now all those companies are targets for potential acquisitions by the behemoths like Datadog, Databricks, and other candidates. The larger companies are not just showing a greater willingness to acquire—they’re doing a lot more of it this year.

[Matthew Lynley @mattlynley]

In the later 10s/early 20s there was practically a venture mania for companies in the orbit of Snowflake and Databricks, with some (like Dbt, Weights & Biases, and so on) reaching lofty valuations. But most you’d talk to would consider it massively overcrowded (and overfunded).

[Matthew Lynley @mattlynley]

Almost got it!

[Gary Marcus @GaryMarcus]

I really have to question anyone who claims that Deep Research is fantastic, when I see so many reports like this.

[Matthew Lynley @mattlynley]

Good bot

[Kyle Russell @kylebrussell]

@atroyn not dealing with getting my kid in a daycare/good school

[Minn @minney_cat]

San Francisco is the most important city in the world, especially true right now.

[Josh Constine 📶🔥 @JoshConstine]

@minney_cat

[Matt Rosoff @MattRosoff]

Some personal news: TechCrunch and I have decided to part ways. It's been a learning experience and I met a lot of great folks there. If anybody is looking for an experienced newsroom leader with a history of training up young reporters to break news, give me a shout.

[martin_casado @martin_casado]

Before you quit .... if you're a good full stack JS/TS dev. Or an SRE. Or a head of engineering. Or a front end dev. Or a designer. Or Python dev. Or back end dev. .... or or or ...

Please apply to the many open SWE reqs in our portfolio! There are so many!

[Eric Newcomer @EricNewcomer]

join newcomer as a senior reporter. all the thrill of going indie and being at the frontier of the media industry

... without the financial risk!

$200k-$300k

[Walter Hickey @WaltHickey]

The book just dropped in Audiobook format, read by the wonderful @JonathanToddRos, it's great listening!