Modeling physical AI risk
San Mateo, CA
Joined September 2025
- Tweets9
- Following27
- Followers73
- Likes36
Valgo retweeted
"vibe coded" is too broad, we should adopt self-driving car terminology instead:
"I just L2'd this tool, what do you think?"
"This site is definitely an L5"
L0: No assistance, done by hand
L1: AI helps you do it
L2: You and AI do it together
L3: AI does it, you actively review
L4: AI does it, you check the result
L5: AI does it, you believe in the ~~vibes~~
We deserve to differentiate "slop" from "AI made me more productive"
—
An L1 post.
Turns out it depends on the city.
humanbaselines.com/analysis
We couldn't find a single place where stakeholders could answer that for their own operating domain and decided to build and publish it ourselves.
The human crash baselines tool covers urban areas and highway trucking corridors, and extends peer-reviewed AV benchmarking methods to heavy-duty trucking. Every public data source and methodological choice is cited and adjustable.
Explore it at: humanbaselines.com
Estimating a crash rate seems simple: crashes divided by miles driven. In practice, the resulting crash rate changes drastically depending on the decisions made in how we define "crashes" and "miles driven."
The domain the AV operates in matters. The right baseline for a robotaxi in a city isn't the right one for an autonomous truck on a highway, and choices like these can shift the number meaningfully.
We independently recreated @Waymo's human baselines, extended them to more cities and trucking routes, and made the tool public. It's meant as a shared reference that's open to contribution, discussion, and critical feedback, which we greatly encourage.
• Explore the tool: humanbaselines.com
• Learn why we did this: valgo.ai/news/humanbaselines
• Learn technical details on how: valgo.ai/pdf/valgo-humanbase…
Valgo retweeted
AVIA is proud to welcome @ValgorithmicInc as the newest member of the organization, bringing new AV expertise to our membership.
Valgo helps AV companies find rare failure scenarios in simulation faster and at lower compute cost, accelerating safety validation before deployment.
Interested in learning about AVIA membership? Visit theavindustry.org/join to learn more.
Valgo retweeted
Insurance runs on billions of historical claims. Autonomous systems have almost none. Whoever prices that risk unlocks a multi-trillion market—why we invested in the exceptional team at @valgorithmicinc, built by @mossr_jl, Sydney Katz & Jon Qian out of Stanford's @aiprof_mykel
Valgo retweeted
Valgo (@ValgorithmicInc) is the risk quantification layer for insuring physical AI.
Since historical claims data doesn’t exist for autonomy, Valgo turns simulations into loss estimates that insurers need to price coverage.
Congrats on the launch, @mossr_jl, @Sydney_Katz, and @JonItoQian!
ycombinator.com/launches/Pde…