Physics & Domain AI
The thesis that AI models trained on physics and deep domain data — surrogate solvers, domain foundation models, physics-informed ML — can attack the computational core of engineering software, not just its edges.
370
active companies
$16.6B
tracked funding
3.65
avg. quality score
$8M
median raise (296 funded)
2020
median founding year
1,145
verified customer relationships
Membership: this thesis appears in the company’s multi-select Investment Thesis field — a company can express more than one thesis. 176 of the 370 companies here also appear in other thesis cohorts. Active companies only; exited companies are excluded.
The Thesis
This is where the money is. Physics & Domain AI carries the largest tracked funding pool of any thesis — roughly $15.8B across 343 active companies in the August 2026 snapshot — ahead of the much larger System of Record cohort. Capital concentrates here because this is the one thesis that attacks the solver core, the part of engineering software with the deepest technical moats and the highest gross margins. Applied Intuition, PhysicsX and Augury lead the ranked list; the funded half of the cohort has a median raise around $7M, but the top carries multi-hundred-million rounds.
The incumbents noticed. The Synopsys–Ansys combination and the wave of simulation consolidation that followed are the establishment’s answer to surrogate models that run orders of magnitude faster than classical solvers on narrow problems. Exits confirm the interest from the other side: this cohort has produced 21 tracked exits, including acquisitions by Siemens and Mistral. Watch the investor column too — NVentures, NVIDIA’s venture arm, shows up here more than almost anywhere else in the database, a distribution-and-silicon bet on physics AI as a compute workload.
Two footnotes worth keeping. First, Germany is the number-two geography, consistent with the broader finding that European engineering AI is undercovered relative to its substance. Second, 133 of these companies also carry the System of Record Enhancement thesis: surrogate models increasingly ship as enhancements to existing toolchains rather than as standalone solvers, which is how a technology thesis turns into a distribution question.
Figures reflect the ThreadMoat dataset as of September 2, 2026. The dataset updates continuously; the numbers in the sections below recompute with every data sync.
Incumbents & Evergreens
Platform incumbents
Established vendors whose segments overlap this cohort most heavily, ranked by shared-segment weight.
Dassault Systèmes
3DEXPERIENCE: CATIA, ENOVIA PLM, SIMULIA, DELMIA manufacturing & operations, Quintiq SCM, BIOVIA R&D
$7.4B
CY2025
Siemens Digital Industries Software
Xcelerator portfolio: NX CAD/CAM, Teamcenter PLM, Simcenter, Opcenter MES/MOM, COMOS, Solid Edge, Altair (2024), Mendix low-code, Dotmatics R&D (2024)
$5.8B
FY2025
Hexagon AB
Metrology, manufacturing intelligence, digital reality, plus Octave Design Intelligence carveout
~$2.5B
CY2025
Autodesk
Fusion 360 CAD/CAM, AutoCAD, Inventor, Revit BIM, Vault/Fusion Manage PLM, MaintainX SCM (acquired 2025)
$7.2B
FY2026
Rockwell Automation
FactoryTalk, Allen-Bradley, Plex MES — industrial automation and smart manufacturing
~$2B
FY2025
AspenTech
Aspen HYSYS, DMC3, Mtell — process optimization, asset performance, supply chain for process industries
$1.1B
FY2024
Synopsys
Full-flow EDA, semiconductor IP, plus ANSYS simulation/multiphysics (acquired 2025) and Granta materials data
$7.1B
FY2025
MathWorks
MATLAB, Simulink — model-based design, controls, simulation, data analytics for engineering
$1.5B
CY2025
Evergreens — durable private players
Long-established, privately held, revenue-led companies that never exit — a different competitive threat from the platform vendors above. The named evergreen roster for this thesis is a dashboard dataset.
Trends Shaping This Thesis
EDA–Simulation Convergence
The Synopsys–Ansys deal and the Keysight–ESI follow-on restructured the simulation market in under two years. Consolidation at the top is the incumbents’ answer to AI-native solver challengers.
The AI Surrogate Shift
AI surrogate models now deliver order-of-magnitude speedups over classical FEA/CFD solvers on narrow problems. The open question is whether they displace solvers or get absorbed as a feature.
European Engineering AI Is Undercovered
The engineering-AI story is told as American, but Germany, the UK and France carry a large share of the serious companies. The EU AI Act is proving to be a moat for compliant vendors, not a barrier.
Full trend analysis with evidence and confidence grading is in the quarterly ThreadMoat report.
Top 3 Startups
Ranked by Weighted Startup Quality Score — the site-wide composite of market opportunity, team execution, technology differentiation, funding efficiency, growth, industry impact and moat. Ties break on verified customer relationships. Scores, funding and stage per company are dashboard data.
Applied Intuition
Augury
367 more ranked companies in this cohort
The full sortable list with score components lives in the ThreadMoat dashboard.
Funding Profile
Latest funding round across the cohort. 296 of 370 companies have disclosed funding above zero; the median total raise among them is $8M.
Denominator: all 370 active companies in this cohort. "Bootstrapped" means no disclosed outside funding, not missing data.
Industries Served
Industries Served is multi-select — a company serves several industries, so bars sum to more than 370. Field coverage: 366 of 370 companies.
Customer Evidence
1,145 verified customer relationships across 937 organizations, joined to this cohort by record ID — never by name matching. Customer evidence exists for 172 of 370 companies; investor data for 267 of 370.
Denominator: 1145 verified relationships (a customer organization can buy from several startups). Full customer-to-startup mapping is a Strategist-tier dataset.
The complete customer intelligence view — which organizations buy from which startups — is in the dashboard.
Geography
HQ country for all 370 companies in this cohort.
Go deeper on this cohort
Every section above has a full interactive version in the ThreadMoat dashboard, filtered and sortable across the complete dataset.
Frequently Asked Questions
What counts as Physics & Domain AI?
Companies whose core product is an AI model of physical or domain behavior: surrogate models for FEA/CFD, physics-informed neural networks, domain foundation models for materials, chips or vehicles, and ML systems that learn from sensor physics. The common thread is attacking computation, not user interface.
Why does this thesis have the most funding?
Because it targets the highest-margin core of engineering software. As of the latest sync the cohort holds roughly $15.8B in tracked funding across 343 active companies — the largest pool of any thesis — led by companies like Applied Intuition and PhysicsX.
Are AI surrogates actually replacing solvers?
Not yet, and the honest answer is that displacement and absorption are both live outcomes. Surrogates deliver large speedups on narrow, well-characterized problems; incumbents are responding by consolidating and by building surrogate capability in-house. The exits in this cohort show incumbents buying rather than ignoring.
How does this thesis overlap with others?
Heavily. 133 members also carry System of Record Enhancement — physics AI increasingly reaches customers as an enhancement to existing tools — and there are meaningful overlaps with Agentic Orchestration and AI-Enabled User Experience.
Related Investment Theses
176 of 370 companies also appear in other thesis cohorts.
System of Record Enhancement
The thesis that the durable value in industrial AI accrues to companies that enhance the established systems of record — PLM, MES, ERP, maintenance and quality platforms — rather than trying to replace them.
Agentic Orchestration
The thesis that AI agents which execute multi-step engineering and operations workflows — not just recommend actions — become the coordination layer across PLM, supply chain, manufacturing and design tools.
AI-Enabled User Experience
The thesis that AI-native interfaces — copilots, text-to-CAD, conversational tooling, automated drawing and model interaction — can win engineering users faster than incumbents can retrofit their products.
Semantic & Knowledge Layer
The thesis that the connective tissue of the digital thread — ontologies, context layers, product knowledge graphs, machine-readable engineering data — becomes its own product layer rather than a feature of the platforms it connects.
Closed-loop Automation
The thesis that the next wave of industrial autonomy closes the loop — systems that sense, decide and act without a human in each cycle, from robotic work cells to self-correcting production processes.
Augmented Knowledge
The thesis that AI can capture, transfer and amplify expert industrial knowledge — training, tribal-knowledge capture, guided operations — as its own product category.
Methodology
Membership. A company belongs to this cohort when “Physics & Domain AI” appears in its multi-select Investment Thesis field. Theses overlap: 176 of 370 members carry at least one other thesis, so cohort sums across pages exceed the database total.
Active filter. Ranked lists include active company groups only (raisers, bootstrapped, scale-ups, unicorns, corporate carveouts). Acquired, IPO’d and inactive companies appear in the exits strip; stealth and speculative outliers never appear. Unlike the site-wide software-only headline count, thesis membership keeps hardware-led autonomy companies — dropping them would erase the Lights-out and Closed-loop cohorts. Company profile links are only shown for companies in the software-only dashboard set.
Honesty rules. Percentages render only above a denominator of 30; medians only with 10+ funded companies; every breakdown states its denominator and field coverage. Customer evidence exists for 172 of 370 companies; investor data for 267 of 370.
Data vintage. Computed from the ThreadMoat dataset synced September 2, 2026. Figures recompute on every sync.