GoPX is a model-agnostic AI control layer. It turns enterprise documents into structured logic, binds AI outputs to verified rules, and makes every decision auditable — so your AI passes the review legal and compliance already require.
GoPX builds on research funded by NSF and DoD and conducted at ASU's CIPS-AI Lab. Enterprise pilot affiliations forthcoming.
Legal, finance, healthcare, and defense operators need AI they can audit. Prompt-wrapped LLMs can't meet that bar — the reasoning is hidden, the behavior drifts, and swapping models erases the work. GoPX was built for the opposite: logic-based explainability, not prompt engineering.
Controllable. Explainable. Trustworthy. In that order.
Not another LLM. A governance layer that travels with your application — swap the underlying model without losing any of the logic.
Convert messy enterprise language — contracts, policies, regulations, manuals — into structured, editable logic: facts, rules, relationships. Not embeddings. Logic your team can read and change.
Evaluate the structured logic across real datasets. Cluster opinions, surface proven patterns, map your organization's actual values, identify what wins. Multilingual — proven on Arabic; generalizes to Hindi, Gujarati, and any underlying LLM-supported language.
Bind every AI output strictly to verified rules and indexed evidence. Hallucination drops because the model is no longer free to invent — it's constrained to cite. Peer-reviewed: 85% reduction in hallucination and toxicity.
A model-agnostic governance layer that sits on top. Your rules, constraints, and exceptions live here — not in a prompt, not in a fine-tune. Switch from GPT-5 to Claude 4 to Gemini 3 without rewriting a line of logic.
A compliance operator asks: "Can we approve this contract under current policy?" GoPX returns a decision — and the exact clauses, rules, and reasoning steps used to reach it.
Can we approve this vendor contract under our current data-handling policy?
Approve with two redlines. Clauses 4.2 and 9.1 require revision before signature.
Policy §3.1 (cross-border data transfer), §5.4 (subprocessor disclosure), §8.2 (audit-rights retention). All versioned, all citable.
Contract clause 4.2 contradicts §3.1 (source text attached). Clause 9.1 omits §5.4 requirement (gap flagged). Remaining clauses pass.
GoPX is the commercial expression of three decades of research by Dr. Hasan Davulcu (Professor, Arizona State University; Director, CIPS-AI Lab) and his group. ACM, IEEE, NSF, and DoD-funded work underpins every stage.
Peer-reviewed in IEEE Transactions on Computational Social Systems: "Beyond the Black Box: Programmable AI and Explainable Text Analysis." — Trivedi, Çetinkaya, Cowan, Newson, Vlahović, Davulcu (forthcoming).
Founder, GoPX.ai · Professor, School of Computing and Augmented Intelligence, Arizona State University
Dr. Hasan Davulcu is a professor at Arizona State University's School of Computing and Augmented Intelligence, where he directs the Cognitive Information Processing Systems (CIPS-AI) Lab.
Ph.D. Computer Science, Stony Brook University · M.S. Computer Science, Stony Brook University · B.S. Mathematics, Middle East Technical University (METU)
Six IP disclosures across twenty-plus method claims underpin the GoPX stack. Three are the backbone of the Interpretable Human + AI framework — the others extend it.
M26-023P Foundational Symbolic pattern lifting and lowering — the core invention behind the control layer. Translates neural outputs into logic-patterned narratives a human or machine can audit, correct, and re-deploy.
M26-021P Safety Programmable stance-directed AI built on the lifting mechanism. Basis of the peer-reviewed 85% reduction in hallucination and toxicity from unrestricted LLMs.
M26-103P Compliance Practical framework for quantifying bias in LLMs — actionable diagnostics for mitigation strategies, regulatory oversight, and the trust reporting that regulated teams already have to produce.
Disclosure IDs verifiable via ASU SkySong Innovations. Exclusive commercial license to GoPX in progress.
When auditors, regulators, or internal risk committees will scrutinize every output your AI produces. GoPX turns policies, regulations, and playbooks into structured rules that bind the model — every decision ships with its cited rules and source evidence, so the review is a reading exercise, not an investigation.
When your AI has to speak authoritatively across fragmented documentation — contracts, procedures, SOPs, regulatory filings — sitting in different systems and business units. GoPX unifies them into one explainable routing layer so the same control logic governs AI behavior wherever it's deployed, and models can be swapped without rewriting the rules.
GoPX.ai offers production-ready LLM solutions that enterprise AI users can trust. We champion programmable AI — agents that carry their own explanation and proof, can be conversationally programmed with business rules and constraints, and are built to say what they do and do what they say.
While competitors grapple with the challenge of controlling their AI models, GoPX is uniquely programmable, explainable, and trustworthy. We leverage metadata, weak supervision, and human-level ground-truthing to encode business and organizational context, constraints, exceptions, and rules — mitigating the inaccuracies and unpredictable behavior that plague other AI systems.
When AI inaccuracies occur, consequences can be severe: eroded trust, wasted resources, and misguided business decisions. GoPX is designed specifically to make those failures detectable and recoverable — by surfacing the reasoning behind every answer.
Three mechanisms: (1) Meta-awareness — Data + Metadata yields superior AI. (2) Ground-truthing with human-level wisdom and judgement. (3) Transparent reasoning — every PX-AI agent shows its reasoning with explanation and proof.
GoPX is founded by Dr. Hasan Davulcu, Professor at Arizona State University's School of Computing and Augmented Intelligence and director of the CIPS-AI Lab. He has over two decades of peer-reviewed research on programmable AI, agentic workflows, sociocultural modeling, and explainable reasoning — published at ACM, IEEE, and NSF/DoD-funded venues.
The featured paper — 'Beyond the Black Box: Programmable AI and Explainable Text Analysis for Trustworthy Social Intelligence' — is the newest statement of the approach and is forthcoming in IEEE Internet Computing. Until it publishes, the 2024 IEEE Internet Computing paper 'Toward a Programmable Humanizing Artificial Intelligence' (same research group, same venue) is the closest published statement. For partnership and enterprise inquiries, use the contact form below.
A 45-minute session. We run your contracts, policies, or regulations through GoPX live and return structured logic, a decision walkthrough, and an honest read on fit. No pitch deck.