Scorecard/Theorem

Theorem

Data gathering in process

AI-powered formal verification lab making program verification 10,000x faster. Mathematical proofs for AI-generated code correctness.

HQUS
Est2025
Size1-10
EU AI ActLimited Risk
theoremlp.com
Score
41.3 / 100
Evidence
5 items

Developing safety practices - core foundations in place with room for improvement.

Strengths:Technical Safety
Weaknesses:Governance Maturity, Risk Assessment, Regulatory Readiness, External Engagement
Focus Areas
formal verificationcode safetyneural theorem provingai safety

Security Assessment

Security-relevant indicators for vendor evaluation

Security Posture
51
TS-01dim: 62
Red Teaming & Pre-deployment Testing
Adversarial testing before deployment
TS-05dim: 62
Robustness & Adversarial Resilience
Resistance to adversarial attacks
RA-01dim: 40
Sector-Specific Risk Assessment
Risk analysis for deployment context
RA-03dim: 40
Dual-Use & Misuse Risk
Dangerous capability awareness
RA-07dim: 40
Incident History & Track Record
Past incidents and response quality
EE-04dim: 40
Vulnerability Disclosure Program
Bug bounty or CVE reporting process
Incident History
Theorem incident records sourced from AIAAIC Repository and public reporting.
Integration: AIAAIC, OECD AI Incidents Monitor
Third-Party Audits
External audit reports, SOC 2 attestations, and ISO certifications verified where published.
Sources: Company filings, registry lookups
CVE & Disclosures
Known vulnerabilities and security advisories from NVD, GitHub Security Advisories, and vendor pages.
Sources: NVD, GHSA, vendor disclosure pages

Dimension Breakdown

GM
Governance Maturitymedium
Published policies, corporate structure, safety mandate, whistleblowing, executive commitment.
35
1 evidence items
GM-01
TS
Technical Safetymedium
Benchmarks, adversarial robustness, fine-tuning safety, watermarking, model cards, research output.
62
2 evidence items
TS-01TS-08
RA
Risk Assessmentlow
Dangerous capability evaluations, thresholds, external testing, bug bounty, halt conditions.
40
1 evidence items
RA-01
RR
Regulatory Readinesslow
ISO 42001, EU AI Act compliance, GPAI obligations, international commitments, incident reporting.
25
EE
External Engagementmedium
Survey participation, research support, transparency, behavior specs, open-source contributions.
40
1 evidence items
EE-01

Social Impact & Safety Profile

Emerging

Theorem's formal verification and code safety tooling contributes to more reliable AI systems, which indirectly supports social impact goals. However, no explicit social impact policy or measurable societal commitments have been published. The connection to social impact is through technical reliability rather than direct engagement.

formal verificationcode safety
Why it matters for safety

Formal verification provides mathematical guarantees about system behaviour. If applied successfully to AI, it could transform safety from probabilistic claims to provable guarantees.

Civilizational Risk Awareness

1/3

Implied safety awareness through positioning. Formal verification approach suggests interest in rigorous safety guarantees, but catastrophic risk framing is not publicly articulated.

Responsible Scaling Policy

None

No RSP. Research and tooling company. Not a model developer.

Mission Drift Protection

1/3
  • ✓Safety research mission
  • ✓Halcyon portfolio alignment
  • ○No PBC status
  • ○No structural governance mechanisms
  • ○Formal verification tooling could be applied to non-safety use cases

Vulnerability Disclosure

None

No CVD programme. Research and tooling stage.

Safety Reporting

- None

No structured safety reporting. Early stage.

Dual-Use Risk

Not applicable - this company does not develop dual-use AI systems.

Need a detailed report for Theorem?

Subscribe to express interest in indicator-level evidence, peer benchmarking, and regulatory gap analysis - or reach out to request a full company overview brief.

Subscribe for UpdatesRequest a Brief
Scoring methodology v0.1 - View full rubric