Audit
Review one project across the six G0–G5 research gates and seven-field Intention–Behavior Record.
PhygitAi
for ChatGPT
OpenAI · Beyond Phygital
Research-Stage Screening · ChatGPTExplore accountable human–AI decision-making in industrial, product, and physical-digital design.
Free download · No ads or tracking code identified in package review
Designing with Intelligence. Designing for Intelligence.
Explore how AI contributes to design, how human decisions are documented, and what evidence supports product behavior and revision.
Choose how to explore your project
Six guided workflows to examine design evidence, identify issues, and explore opportunities for revision.
Compare
Examine two projects using consistent evidence criteria, without scoring or ranking.
Inspect
Examine project claims, supporting documents, and missing evidence.
Detect
Identify documented discrepancies, evidence gaps, and unclear decision authority.
Explain
Clarify existing PhygitAi assessment findings and their methodological limitations.
IBR
Trace one design decision using the seven Intention–Behavior Record fields.
Conceptual research · Human-directed decisions · Evidence-based screening
G0–G5 research gates
G stands for gate: a research checkpoint for organizing evidence and questions requiring human judgment.
Intention & Assumptions
Clarify the design purpose and assumptions.
AI-Assisted Co-Design
Document AI contributions and human selection.
Authority & Delegation
Identify who can decide, approve, and override.
Physical Realization
Connect the design to implementation and observations.
Disagreement Assessment
Examine differences between intended and observed outcomes.
Accountable Revision
Record decisions, corrections, and retesting.
The seven-field Intention–Behavior Record (IBR)
A focused trace of one design decision and its follow-up.
- Intention
- AI Input
- Permission
- Event
- Outcome
- Decision
- Retest
Security & Responsible Use
Official PhygitAi public-screening packages are intended for legitimate research, education, and design review, without advertising, spam, or unauthorized tracking. The private research backend and unpublished research master are not distributed in these packages.
Static inspection of the current downloadable archives found text instructions, templates, examples, and configuration only, with no executable payloads, configured remote assessment services, or tracking mechanisms identified. This review is limited to the exact current archive checksums in the linked report; it does not verify an installed plugin or the host platform.
Use only trusted releases and share material you are authorized to disclose. ChatGPT, Claude, and Gemini process information under their own policies. Screening outputs require human review and are not empirical validation or certification.
Package checks, SHA-256 checksums & limitations
The package verification report records archive and member checksums, inspected contents, and findings for ChatGPT v0.1.8, Claude v0.3.1, and Gemini v0.1.5. The ChatGPT Complete ZIP includes its referenced assets. A separate earlier update-only ZIP is not the current download. The corrected Claude standalone Skill includes the previously missing notices and corrected layout guidance. Revised-version host import tests remain unverified.
No antivirus scan, runtime or network test, or independent security certification was performed. No guarantee that files are virus-free is made. Changed packages require a fresh review and checksum.
Proprietary Licensing
PhygitAi is proprietary research material by Abbas Sibai, Ph.D. Public screening does not imply an open-source license or permission to redistribute the methodology.
The supplied release archives declare © 2026 Abbas Sibai, Ph.D., all rights reserved, and contain no standalone license file. This page grants no additional redistribution rights. Consult the applicable release terms or contact the author for permission before redistribution. The private research backend and unpublished research master are not included.
The complete screening archive is downloadable here. Existing private-plugin access remains separately restricted by the host platform.
Conceptual research limitations
PhygitAi is a conceptual research methodology under development. Screening outputs depend on the supplied evidence and may contain errors. Missing information must remain explicit; reported claims are not independently verified facts.
AI-generated preliminary alignment reviews are not empirical validation, professional advice, product certification, or endorsement by the framework originator. Human reviewers retain responsibility for decisions.
See the Privacy Policy, Terms of Service, and Support.