Go / No-GoLive
AECO Shield
Is this defensible? Score AI tools, derive Stamp-Safe verdicts, store the audit trail.
Register interest →New research
How AI is being written into — and out of — the insurance behind stamped engineering and design work. What the paperwork says, and what’s being sold on top of it.
Free to read · no sign-up
Independent research
Every recommendation carries an audit trail. Every score is calibrated to the AECO Compliance methodology. The kernel is the moat.
Methodology v3.4 · ACS™ · Stamp-Safe™
What Anthropic is to finance compliance, AECO.digital is to AEC compliance.
Architecture · L0–L6
The work
Shield expresses the AECO Compliance kernel into a decision a licensed professional can defend. Other applications of the same kernel are under consideration — none is announced, dated, or available.
Is this defensible? Score AI tools, derive Stamp-Safe verdicts, store the audit trail.
Register interest →Standards we track
The methodology maps to every framework your E&O carrier, license board, and procurement team will ask about.
BIM · Information mgmt
ISO 19650
DIS public consultation · final expected 2027
AI · Management system
ISO/IEC 42001
First international certifiable AI management standard
AI · Risk framework
NIST AI RMF
Active · 1.0 + GenAI Profile · 1.1 in progress
EU · Enforcement
EU AI Act
Annex III enforcement · Aug 2026 · penalties up to €15M / 3% turnover
DoD · Cybersecurity
CMMC 2.0
Phase 2 begins Nov 2026
US state · AI
Colorado AI Act
SB 24-205 effective 2026 · NIST AI RMF as affirmative defense
Legal · Precedent
ABA Opinion 512
Jul 2024 · guidance on AI in professional practice
Governance · Principles
IBM AI Governance
Industry governance principles
Standards reviewed quarterly
Platform mission
“Which AI tools can your firm actually deploy — without exposing your licence, E&O coverage, or federal contract?”
That question has no quick answer. We built three modules to generate a defensible one.
Score any AI tool across all 7 liability layers using publicly available evidence, vendor documentation, and AEC industry knowledge — no vendor self-reporting.
Goal: ACS score 0–10
Profile your firm's AI readiness across 7 dimensions — from BIM maturity to AI governance — to establish the baseline your FitScore calculation depends on.
Goal: Firm Maturity Profile
Combine the tool's ACS score with your firm profile and project context to produce a ranked, defensible FitScore — with hard-filter blocks for critical compliance failures.
Goal: FitScore 0–10 + Go/No-Go
All three work together to answer one question: is this tool safe to deploy, right now, for this project?
Which account is yours?
AECO Shield has two sides. Firms assess the AI tools they use; vendors get their tools assessed. You pick which at signup.
You're a firm account if you use AI tools on stamped or regulated AEC work — a licensed professional (PE / AIA / CCM / LS), project manager, BIM manager, pursuits lead, or executive.
What this account does
You're a vendor account if you build an AI tool used in AEC and want it independently assessed and listed where firms evaluate tools.
What this account does
You choose your account type when you sign up. Firm and vendor accounts are separate — pick the one that fits how you use the platform.
The public registry
Every AI tool AECO Shield assesses is published to a public registry — score band, go/no-go signal, and Stamp-Safe verdict, readable by anyone. Firms browse before they buy; vendors get discovered; carriers and boards can verify. No login required to look.
Sample AI Tool
by Illustrative Vendor
Illustrative — not an actual registered tool.
Public browsing is free · deep actions (score your own tools, full reports) need an account.
What you get
The ACS score tells you what the vendor documentation actually says about liability — not what the sales deck claims. The FitScore adds your firm maturity and project context to give you a ranked, defensible decision.
Stamp-Safe — a Licensed Professional can place responsible charge on this output
Conditional — use with documented governance conditions
Not Stamp-Safe — do not use on stamped deliverables
ILLUSTRATIVE EXAMPLE
Your AI Tool
7-layer compliance check
Illustrative example — actual scores reflect live assessment data.
The problem
No professional stamp disclaimer. No code edition validation. No human review gate. No insurance coverage data. If something goes wrong on a stamped deliverable — who is liable?
Most AI tools ship without any professional-license disclaimer
Vendor documentation rarely addresses professional liability or responsible charge requirements.
CMMC enforcement begins October 31, 2026
Federal AEC contractors cannot bid on new DoD contracts without verified compliant AI tools.
Procurement decisions are based on vendor claims
No independent, AEC-specific compliance verification tool existed — until now.
Why it matters
The stakes for licensed AEC practice are concrete:
Federal Design-Build with CMMC 2.0
A licensed practitioner selecting an AI tool for a project the contract requires to pass CMMC.
Without a compliance gate
The CMMC gap surfaces mid-project during audit. Contract exposure follows — schedule, remediation, and potential E&O claim under insurers’ AI-exclusion coverage clauses (Verisk CG 40 47/48).
With AECO Shield
Shield’s data-security and compliance gate flags the CMMC gap at project setup. Tool selection is reviewed before contract. The FitScore report attaches to the procurement audit trail.
6–7 figures
E&O defence costs
a single AI-tool misuse on stamped work can trigger
Many
AI deployments fail
on organisational readiness, not tool quality
Oct 31 2026
CMMC hard deadline
non-compliant firms cannot bid on new DoD AEC contracts
51
US jurisdictions covered
AHJ code edition lookup runs automatically across all 50 states + DC
Assess your firm (10 min)
Complete the Digital Maturity Assessment across 7 dimensions — BIM readiness, AI governance, ISO 19650 stage, supply chain alignment, and professional oversight protocols.
Set up your project
Define project type, phase, jurisdiction, and delivery method. The AHJ code edition lookup runs automatically across 51 jurisdictions. Hard compliance filters run before any score is shown.
Get your FitScore
Every AI tool is ranked against your firm profile and project context. Tools that fail a hard-filter check — responsible-charge fit, the vendor's technology E&O and cyber coverage, or your firm's data-security requirements — are blocked before scoring. Remaining tools ranked 0–10.
Digital Maturity Assessor
40 questions across 7 dimensions. Benchmark against your peer group. Identify the governance gaps that block safe AI deployment.
Firm Maturity Profile
IllustrativeD4.5 ceiling rule caps Process & Governance at 4.9 if no named PE reviewer is designated.
Benchmark Comparison
| Dimension | Your Firm | Peer Median | Top 25% |
|---|---|---|---|
| D1 Data | 6.2 | 5.8 | 8.1 |
| D2 BIM | 7.8 | 6.4 | 8.9 |
| D3 AI | 4.1 | 4.9 | 7.2 |
| D4 Gov | 5.5 | 5.2 | 7.8 |
| D5 Talent | 6.0 | 5.5 | 7.6 |
| D6 Scale | 5.2 | 4.7 | 7.0 |
| D7 Supply | 4.8 | 4.3 | 6.5 |
What this means for FitScore
A firm with D3 AI Readiness below 5.0 will receive a reduced MaturityMatch component for any AI-heavy tool. The platform flags this before the deployment decision — not after the tool is live on a project.
KPI Progress — Illustrative 3-Year Trajectory
Year 1 Baseline → Year 2 → Year 3 Target
Track your firm's compliance maturity year-over-year. The platform generates a printable KPI Executive Summary with recommendations — ready for board review or ISO audit preparation.
Register interest in the report →Scoring framework
The ACS scoring framework covers every liability surface that matters in AEC professional practice. Layer weight reflects relative exposure — the brightest arc carries the most risk.
Platform architecture
C1 scores the tool. C2 profiles your firm. C3 combines both with your project context to produce a ranked, defensible FitScore.
The questions AEC practitioners actually face — answered.
“Our AI tools need to pass CMMC by October 2026. Where do we stand?”
The platform scores every tool in your stack against CMMC 2.0, SOC 2 Type II, and NIST AI RMF. Hard filters block tools whose vendors lack adequate technology E&O and cyber coverage, or whose data-security posture falls short, before FitScore is calculated.
L5 Cybersecurity · Hard Filter Pipeline
Outcome: Tools in your stack cleared or blocked before the CMMC deadline.
“We're a European AI vendor. Are we ready to sell into the US AEC market?”
Shield’s US Market Readiness assessment maps your tool against what US firms actually check before procuring — responsible-charge fit, technology E&O and cyber coverage, data-security posture (SOC 2), and state AI regulations. Gaps are surfaced with a prioritised remediation roadmap.
US Market Readiness · ACS Registry
Outcome: Six-month US market entry research compressed to a prioritised roadmap.
“We used AI to assist structural calculations. Can we stamp them?”
The Stamp-Safe classification specifically answers this. It checks professional disclaimer disclosure, human-in-the-loop checkpoints, and the vendor's professional stamp compatibility declaration.
L1 Responsible Charge · Q21b · Stamp-Safe
Outcome: PE licence protected. Stamp decision documented before it is made.
The positioning
A licensed professional doesn't need help finding information. They need help proving the information they used was defensible.
| The question that matters | Asking AI directly | AECO.digital |
|---|---|---|
| Tool defensibility for stamped work | ✗Guesswork — the AI itself doesn't know | ✓22-question ACS rubric across 7 domains, weighted by professional + legal exposure |
| Statistical confidence floor | ✗No threshold — same tone regardless of certainty | ✓DCS thresholds: GREEN ≥88% · AMBER 68–87% · RED <68% · transparent, versioned calibration framework |
| Output insurers will accept | ✗Chat transcript — not accepted by E&O carriers | ✓Audit-grade PDF with methodology version, evidence citations, L1 attestation, 7-year retention |
| Jurisdiction-specific code awareness | ✗Generic answer — no AHJ context | ✓51 US jurisdictions (50 states + DC) · AHJ code edition lookup runs automatically |
| Source citation requirements | ✗Citations optional · may hallucinate · cannot be filtered by jurisdiction | ✓Citations grounded in the L2 standards library · applicability-tagged (topics, disciplines, use cases) |
| Liability transfer test on tool ToS | ✗May miss aggressive vendor ToS clauses | ✓Q21b: dedicated ToS liability-transfer test · fires warning banner at score ≤ 2 · 0.85× MaturityMatch penalty on CD/Construction phases |
| Audit trail with longitudinal memory | ✗No memory between sessions | ✓7-year immutable record · methodology v3.4 stamped on every artifact |
| Longitudinal firm posture tracking | ✗No firm posture · no maturity tier · no trend | ✓FMP profile tracks D1-D7 over time · 5 maturity bands (Emerging / Developing / Established / Advanced / Leading) · exportable trend reports for E&O carrier or reviewer review |
Ask a general-purpose AI to evaluate an AI tool and you get an optimistic answer. Run the same tool through AECO Shield and Q21b reads the actual liability-transfer language in the terms of service — the specific question that matters before a licensed professional places their stamp.
AI is excellent at answering questions. AECO.digital answers the specific question that matters before a licensed professional places their stamp.
Use Claude to discover tools. Use AECO Shield to decide if you can safely deploy them.
Tool scores are never influenced by vendor relationships or payments. Every score is evidence-based research from public sources.
All scoring formulas authored by a practising structural engineer with PhD, PE, PMP, and IWE credentials. Engineering rigour, not marketing claims.
Payment covers the assessment process — not the outcome. A tool that fails the 7-layer framework receives a low score regardless of what was paid. Editorial independence is non-negotiable.
Research & preview
AECO.digital is an independent research and methodology project. It is not for sale today — the methodology is published, the registry is open to browse, and you can register interest in what comes next.
The compliance workflow the methodology describes — tool scoring, sign-off, audit artifacts — is in active research and preview.
Register interest →How tools are assessed against the 7-layer framework is published in full. Registry listing is not open for purchase.
Register interest →