Safeguarded AI for Supply Chains

Developing state-of-the-art AI for safety-critical supply chains, with the University of Oxford
April 1st, 2025
HASH
Safeguarded AI for Supply Chains

The Safeguarded AI Project

HASH is leading a series of work for the UK's Advanced Research + Invention Agency (ARIA), supporting the research and development of safe AI solutions for chemical and biopharmaceutical supply chains.

Oxford Partnership

The “Safeguarded AI for Logistics & Supply-chains” (SAILS) project, led by HASH, is being undertaken in conjunction with researchers from the Department of Computer Science at the University of Oxford.

Jointly, HASH and Oxford are developing state-of-the-art, AI-based solutions to challenges that exist in chemical and biopharmaceutical supply chains, modeling processes such as new product introduction, network optimization, safety stock sizing, and temperature excursion handling.

Nobuko Yoshida, Oxford’s Christopher Strachey Chair of Computing, said, "We’re excited to be working with the team at HASH, building on their experience developing trustworthy AI for safety-critical industries, and combining it with our research."

Research Overview

Prof. Yoshida is best-known for her development of multiparty asynchronous session types, which help describe communication protocols used in concurrent and distributed systems. As part of her work on this project she will be developing the concept of probabilistic session types. She is joined by fellow Professor of Computer Science at the University of Oxford, David Parker.

Prof. Parker’s expertise lies in probabilistic verification, and he is one of the creators of the PRISM tool for formal modeling and analysis of stochastic and probabilistic systems.

Senior Research Associate, Dr Adrián Puerto Aubel, an expert in process modeling using Petri nets, joins them. Describing his work, Adrián explains, "Petri nets are formal models of distributed systems. They were conceived as models for parallel computing, but they are extremely well suited to describe supply chains and business processes. Extensive theory — developed over decades — provides means to derive models automatically from data (a technique known as "process mining"), as well as a myriad of verification techniques. Their success is due to the great flexibility they provide for modeling systems and their blend of mathematical rigor and intuitive visual representation. As a result they lie at the core of process science. They support the full workflows of business process and supply chain management from design and compliance, risk and weakness analysis, up to online monitoring of business activity. Petri nets bring business processes as first class citizens into the realm of computer science."

The SAILS project forms a part of ARIA's "Mathematics for Safeguarded AI" program, led by David "davidad" Dalrymple (Program Director) and supported by Yoshua Bengio (Scientific Director).

davidad explains, “AI could unlock transformative improvements in our critical infrastructure, but right now adoption is limited because without ironclad safety assurances, we risk unintended and damaging consequences. Our goal is to prove that it’s possible to develop AI with quantitative safety guarantees, and that this could unlock significant economic value for the UK.”

Broader Context

ARIA is an R&D funding agency created to unlock technological breakthroughs “that benefit everyone”, closely aligned with HASH’s mission to “help everybody make the right decisions”. ARIA is sponsored by the United Kingdom's Department for Science, Innovation and Technology, and the organization funds teams of scientists and engineers to pursue research "at the edge of what is scientifically and technologically possible."

This work builds on HASH's experience to date working across life sciences, and atop its existing data extraction and simulation modeling technology.

Leah Pickering, former Global Planning Process Lead at AstraZeneca, joins HASH in support of this project.

Get Involved

Chemical and biopharmaceutical companies interested in benefiting from the technology developed as part of this program are invited to reach out to learn more about getting involved, by emailing Leah Pickering ([email protected]).

Organizations in other industries similarly interested in high-trust, “safe” AI supply chain technology are invited to directly contact HASH’s CEO, Dei Vilkinsons ([email protected]).

Create a free account

Sign up to try HASH out for yourself, and see what all the fuss is about

By signing up you agree to our terms and conditions and privacy policy