DRONE DATA & AI TOOLS

How can data and AI tools become more relevant to solving local social challenges?

Data analysis is essential to supporting policies in the Global South, informing sound policy development, and ensuring overall societal well-being. Yet, tools for the interpretation and use of data are disproportionately developed for and by engineers from the Global North.

The results of this are tools and hardware that are poorly adapted for the needs of the Global South and create dependency on foreign companies. For instance, drone analysis tools are often reliant on continuous internet connections and require significant data processing power, which is not always available in rural contexts. Furthermore, for various reasons – including economy of scale and profitability – current data analysis tools are designed to address generic problems; e.g. photo album organization, satellite imagery classification, or web data aggregation, rather than the specific problems faced by communities in the Global South. When the tools are relevant, the cost of using the platform or hosting the data is a barrier to entry for users in developing nations. Finally, advanced image analysis tools require large training and validation samples, which are often kept by large international technology firms. When libraries are made public, these training samples do not necessarily apply to local challenges. The consequences are forgotten local and offline communities in the Global South without access to the full promises of drone and robotics development.

Creative solutions to adapting drone and AI technology to the needs and contexts of the Global South are urgently needed. The goal of this challenge is to solicit ideas and solutions which will provide the analytical tools required to make drone and AI data actionable in Global South.

Help us make data & AI tools part of the solution, not the problem!

Last Data Mile

How can we turn data into action by bringing back drone data and insights to beneficiaries?

Once drone data has been processed into data products and analyzed to generate insights, turning these insights into action and positive outcomes for all beneficiaries is the last hurdle to overcome. For example, how can local authorities apply risk recommendations from drone data analysis to their community? Or how does a farmer translate a vegetation index into drought areas of fields and adapt irrigation? These are just 2 of hundreds of examples. Finding easy and intuitive ways for beneficiaries to visualize data and turn it into action will have a positive impact on the system, thus closing the feedback loop.

Current solutions mainly don’t take into consideration the needs of local beneficiaries. Most beneficiaries don’t live in a tech oriented environment, nor are they data experts. Many beneficiaries have limited data handling capacity and live in low-tech environments with limited or no access to connectivity and bandwidth.The sheer size of drone data and the fact that many data products and analysis tools are hosted online or ask for the use of high-performance equipment make it difficult to bring back data and insights. Finally, drone data is only a piece of the puzzle to solve local social challenges and needs to be integrated with other data sources to create actionable insights.

The goal of this challenge is to find solutions that close the feedback loop in an intuitive and low-tech way that is fitting for non-data experts.

Help us to tackle the Last Data Mile and bring back the data and insights they generate to the beneficiaries!

DRone & data ethics

How can drones and data be used ethically to solve local problems?

Drones pose a number of ethical challenges when used for social good. The data collected using drones may put people at risk by exposing personal, community and demographically identifiable information, for example. This explains why informed consent is a foundational ethical principle. However, gaining consent presents logistical and practical challenges, especially in emergency contexts when time is of the essence. (It is worth emphasizing that satellite companies are not required to gain informed consent even though they collect massively more data).

Another ethical problem posed by drones is data trawling, which consists of collecting as much as data as possible regardless of need for said data. Only the absolute minimum amount of data that is necessary to address a specific problem must be collected. All other data should be spared.

In addition, the use of drones in social good projects face a range of structural inequalities in terms of market access for local drone experts. Costs, regulations, certifications and outright discrimination all serve as important barriers to entry for local drone experts to participate in social good projects within their own countries. There are concerns that international organizations are increasingly monopolizing the drone space, making it more difficult for local players to solve local problems. This too is unethical and exacerbates inequality and the digital divide.

These ethical problems are illustrative examples. There are of course other ethical issues out there like AI bias, for example. As such, we invite participants to propose solutions to other ethics issues as well.

Help us integrate ethics as part of every solution!