WHO WE ARE PROJECTS GET INVOLVED

why we care

according to the UN, in 2016 the annual value of illegal wildlife trade is estimated to be worth USD 7–23 billion

“Wildlife trade escalates into a crisis when an increasing proportion is illegal and unsustainable—directly threatening the survival of many species in the wild.” - WWF

Latest Projects

Microsoft UK’s SEEKER

 SEEKER AI by Microsoft Project Carriers

Microsoft UK,Heathrow Airport, The Royal Foundation’s United for Wildlife programme

Beneficiaries

wildlife, humans

Users

people working at transit points such as airport, port (transport sector)

Need

protection of the animal ecosystem, wildlife trafficking rates fell sharply during COVID

Principles

make it impossible for traffickers to transport, finance or profit from illegal wildlife products, multispecies AI that scans luggage and cargos and automatic alert enforcement agencies upon detection of illegal wildlife

Main Technologies Involved

simple installation at airports, Azure-based technology, machine learning, IoT, DevOps tools, scanned up to 250,000 bags despite few data on illegal wildlife trafficking

Sources

https://www.securitymagazine.com/articles/96564-ai-tracks-illegal-wildlife-trafficking-at-heathrow, https://news.microsoft.com/en-gb/2021/11/18/first-of-its-kind-multispecies-ai-model-to-detect-illegal-wildlife-trafficking-is-ready-to-roll-out-to-airports/ https://www.microsoft.com/en-gb/ai/ai-for-good

Coalition to end wildlife trafficking online

 Coalition to End Wildlife Trafficking Online Project Carriers

tech companies, WWF, TRAFFIC, IFAW

Beneficiaries

wildlife

Users

companies, law enforcers

Need

illicit wildlife trade results in unprecedented declines in a variety of species population, globally BLA countries have been recorded to have, prohibition of wildlife consumption

Principles

companies are a key force in combating cybercrime, co-operation across business sectosr can help with implementation

Main Technologies Involved

simple installation at airports, Azure-based technology, machine learning, IoT, DevOps tools, scanned up to 250,000 bags despite few data on illegal wildlife trafficking

Sources

https://www.worldwildlife.org/initiatives/wildlife-conservation https://www.worldwildlife.org/stories/tech-companies-remove-or-block-11-6-million-listings-for-prohibited-wildlife-and-products

Wildlife Sentinel

 mobile App Wildlife Sentinel Project Carriers

Crime Stoppers International, USAID Reducing Opportunities for Unlawful Transport of Endangered Species (Routes), International Air Transport Association, NGOs

Beneficiaries

wildlife

Users

staff from airlines, at airports, aviation companies

Need

Plane Sight Report: trafficking in aviation was found in 136 countries, increase of 40% in wildlife seizures in air transportation, Combat against wildlife trafficking in aviation, immediate use once the app launched

Principles

an application for employees in the aviation industry to anonymously report suspicious activities, Form: flight details, location, details of the people involved Several languages are available: currently English, Spanish and Portuguese

Main Technologies Involved

mobile app to help stop the trafficking of wildlife in the aviation industry

Sources

https://www.traffic.org/news/new-mobile-reporting-app-is-helping-combat-corruption-and-wildlife-trafficking-in-the-aviation-industry/

Predictive Modeling of High-Risk Taxa

 Project Carriers Predictive Modelling Example

WWF & TRAFFIC, tech companies

Beneficiaries

wildlife, humans

Users

authorities, companies, WWF

Need

prevention of potential pandemics in light of COVID-19,link between health, wildlife exploitation and illegal traffic might accelerate

Principles

identification of emerging trends in sale of species (online and offline) that may be in risk of zoonotic transmission and the potential hotspots

Main Technologies Involved

predictive modelling tool: machine learning, big data system uses a mix of online search and data from social media, market research and other data sources

Sources

https://www.worldwildlife.org/initiatives/traffic

On the News

Microsoft UK’s SEEKER

How Microsoft developed the first AI that recognises illegal trafficked animals in luggage

In collaboration and partnership with the UK Border Force Cite, UK security and screening technology vendor 'Smiths Detection' and other organisations, especially airports, Microsoft's broke through with their AI model to recognise smuggled animal parts in luggage. This is part of their Microsoft "AI for Good" project.

How can image recognition support the combat against the traffic and sale of endangered species both in 'hot spots' and in e-commerce?

Image recognition and machine learning can raise awareness of technology, which may play a major key role in fighting criminal networks for trafficking of rare species. Given its popularity in Asia, Europe is often the hot spot for transit of poachers. Hence, it involves several stakeholders such as those transport, finance, non-governmental organisation and law makers. In collaborative efforts the sharing of information and expertise supports an internationl approach in dismantling the criminal network. Mircosoft has developed the first successful machine learning to recognise trafficked animal parts. It is truly 'AI for Good'.

Users

transport sector, government & non-governmental organisations, policy makers

Key Features

1. detect trafficked animals
2. understand the destination
3. track poachers
4. reduce trafficking of animal

UX Storyboard

UX Storyboard for Animal Trafficking

General Principle

SEEKER is the first AI model which can detect illegally trafficked wildlife in luggage.

Technical Overview

Computer vision & image recognition make up the core part of this supervised machine learning. Microsoft's algorithm uses images of animals and specific body parts. the system does a 3D scan of the luggage, generated by Smiths Detection’s X-ray scanners. Further, the image is rotated for a full view. Any unrelated objects such as shoes are filtered while the animal object is matched with the image the algorithm was trained on. Within two months, the algorithm can be trained on any species using a restricted dataset.

Added Value

Better detection rates for scanning illegal trafficked animals may be the main feature, but from there authorities can track the routes and journey of smuggling, leading to a better fight against illegal wild animal trafficking.

Get involved

Coalition to end wildlife trafficking partners