Why Zig Zag?
Human Truth in an AI World: Where Technology Meets Meaning
​
While AI can process vast amounts of data, it fundamentally cannot understand context, nuance, or cultural meaning—what we call "dog whistles" and narrative frames. ZigZag bridges this critical gap by combining rigorous mixed-method research with AI capabilities. We don't just label content; we map the structure, spread, and real-world impact of hostile narratives. This means you get defensible, audit-ready analysis that stands up to regulatory scrutiny, board-level questioning, and ethical review—because real decisions need human truth, not algorithmic guesswork.
​
For Corporate clients: Meet MRM compliance and ESG requirements with transparent, explainable AI systems.
For Government clients: Achieve AI assurance that aligns with UK frameworks and withstands operational pressure.
For Civil Society: Deliver donor-ready, ethically sound analysis that protects your mission and reputation.
Confidence as Your Competitive Advantage: From Risk to Strategic Asset
​
In high-stakes environments—whether you're managing financial compliance, national security threats, or human rights investigations—uncertainty is expensive. ZigZag transforms AI from a black-box risk into a confidence multiplier. Our Test & Evaluation services and Demystifying AI Training give your teams the assurance to act decisively, the transparency to defend their decisions, and the insight to stay ahead of evolving threats. We turn model validation from a compliance checkbox into strategic intelligence.
​
The outcome: Boards trust your AI. Regulators accept your methodology. Your team makes faster, bolder decisions with full accountability.
Born from Research, Built for Reality: Credibility You Can Deploy
​
ZigZag didn't emerge from a corporate strategy deck—it was forged in the field, studying hate speech, hostile narratives, and the collision between free expression and harm across platforms and languages. Our methodology has been tested against state-of-the-art AI and validated through real-world operations. We understand that context is everything, and we've built our services to handle the messy, contested, high-pressure scenarios where definitions fail and stakes are highest.
​
What this means for you: No theoretical frameworks that collapse under pressure. No one-size-fits-all tools. Just proven, pragmatic solutions designed for the unique demands of Financial Services, Defence & Security, Policing, NGOs, and Human Rights organisations.
Meet
The Team
Meet the team who turned years of field research into the methodology that powers ZigZag - combining expertise from academia, intelligence, compliance, and human rights to solve problems where AI alone fails.
Dr Stephen Anning
Read Bio
Rebekah Lattin-Rawstrone
Read Bio
Our Services
ZigZag combines graph-based natural language processing with narrative analysis methodologies grounded in social science. Our high-integrity analytics toolkit detects the structure and spread of hostile narratives, recurring storylines that target identities, institutions, and norms.
Beyond AI: Human-Level Understanding
Demystifying AI
Training
We provide training sessions to demystify artificial intelligence and ensure the technical feasibility of your strategic ambitions.
Test & Evaluation Services
We offer test and evaluation services to assess the integrity and performance of your current AI systems.