What AI do we build?
We build AI that reduces uncertainty in operations: it detects problems early, forecasts what is likely, and removes manual document work, always inside accountable systems where people decide. Our AI products are developed through ClarityAILab, our AI lab, with governance built in.
Early warning and anomaly detection
Our early warning algorithm detects sudden departures from established patterns. It analyses historic data to establish what is normal, including random variation, and flags changes that matter.
Applied to mortality data, it identified a significant spike in deaths in early January 2020, weeks before the first official COVID-19 case was reported in the UK. While anecdotal reports hinted at earlier outbreaks, the algorithm gave quantitative evidence of the virus's early impact. Read the case.
The same method can monitor sales, production processes, supplier performance or any operation where catching a change early reduces risk.
Optimization that searches wider
Most planning tools converge on the first good answer they find. Our stochastic optimization, enhanced by genetic algorithms, keeps exploring alternative assumptions, so the plan it returns is robust rather than just locally optimal. See decision support.
Document AI for supply chains
Supply chain decisions often wait on documents: invoices, packing lists, spreadsheets, tenders. MaiMai, developed through ClarityAILab, turns them into structured data and runs sourcing and purchasing workflows with approvals and audit trails.
How we keep AI accountable
- AI supports human decisions; it does not replace them.
- Outputs are traceable and explainable to the people accountable for them.
- Data stays under the owner's control.
- Systems must work in real operations, not only in tests.
More on our approach in Governance.
Questions people ask
What is an early warning algorithm?
An early warning algorithm learns the normal pattern and random variation in historic data, then flags sudden departures from that pattern. Ours identified a significant spike in deaths in early January 2020, weeks before the first official COVID-19 case was reported in the UK.
Where can anomaly detection be used in a business?
Anywhere early detection reduces risk or opens an opportunity, including sales, production processes, supplier performance, logistics delays and fraud. The same method applies wherever there is enough history to define normal.
Does Building Resilience's AI replace people?
No. Our AI is designed to support human decision-making, not replace it. It highlights patterns, risks and options; accountable people decide, with clear boundaries, oversight and audit trails.