← Back to AI for Good

Upstream

A Fieldcraft Studios × Braidr Research Initiative

Mapping the gap between what UK water users care about and the invisible threat of antimicrobial resistance — across 10 communities and 3 platforms.

Loading… March 2026 TikTok 10% · Instagram 100% · Reddit 80%
About Upstream

Upstream is a joint research initiative from Fieldcraft Studios and Braidr, built as part of Braidr's AI for Good programme. The name points to where the problem starts — contamination enters UK waterways upstream, and that's where communication needs to go too. The project uses AI-powered audience intelligence to map the communities most at risk from antimicrobial resistance in UK waterways, and to understand how to actually reach them. Everything here is public, free, and designed to be used. This is living research, not a finished report.

Executive
Summary

Loading analysis data…

—
Communities Analysed
—
Posts Analysed
—
Creators Tracked
—
Total Engagements
—
Combined Reach
Section 01
Comparative
Matrix
The core deliverable: seven strategic dimensions mapped for each UK water user community. Each row summarises the community's relationship to water, awareness of invisible threats, trust dynamics, and the messaging pathways most likely to land. Click any community name for the full deep-dive.
Loading community data across all platforms…
Community Opportunity SizingHow big? How active? Awareness GapAMR knowledge vs. adjacent concerns Platform IntelligenceWhere, what formats, who? Value MappingCore drivers, tensions Entry PointsGaps to fill, narratives to join Authenticity CalibrationWho do they trust and reject? Message HooksFraming that lands
Section 02
Cross-Community
Insights
Patterns observed across all communities — the shared blindspots, common barriers, and universal entry points that should inform any cross-community AMR communications strategy.

Methodology & Approach

Data Collection

Social media content was collected from TikTok, Instagram, and Reddit using community-specific search queries and hashtag ecosystems. All content was filtered for UK relevance to ensure analysis reflects the domestic landscape.

Analysis Pipeline

Content was analysed using large language models guided by a research-grounded system prompt encompassing AMR science, risk communication research, and UK regulatory context. The LLM produced structured outputs following a defined analytical schema.

Research Grounding

Analysis is informed by published academic and government sources including the Beach Bums study (University of Exeter), Environment Agency pilot surveillance data, Wellcome Trust communication research, and the UK AMR National Action Plan 2024–2029.

Matrix Construction

Per-platform analyses were synthesised into cross-platform community profiles, then assembled into the comparative matrix. Each of seven dimensions maps a specific strategic question relevant to campaign design.