A Conversation with Gayan Peiris, Head of Data and Technology at UNDP
Edition #81 UNDP's Head of Data and Technology talks AI transformation, systems-thinking, and defining digital in one of the UN's largest agencies...
This week, interweave sat down with UNDP’s Gayan Peiris to discuss how digital has shaped the UN’s development agency, including what it took to become the UN’s first “digital-first” body and what development looks like in a digital context. If this piece is of interest, you may also enjoy our conversation with Yolanda Martinez, UN ITU’s Head of GovStack, and Barbara Ubaldi, Head of Digital at the OECD.
Since the beginning of the year, we have been occasionally sharing some other resources in the digital government space that we are reading, and are huge fans of. This week, we wanted to shout out our friends at the Civic Tech Field Guide. Their weekly Friday newsletter shares gems from across the global field of civic tech. Each issue features new civic and GovTech product launches, upcoming funding deadlines, and hand picked top jobs. Be sure to check it out!
Gayan Peiris’s UN career has coincided with something of a digital revolution. When he joined the UN in 2006, it was as a web and content management consultant, at a time when “digital” was still in its infancy. IT projects were “random and disconnected” rather than representing “a holistic approach or strategy” – and the scope of a “digital project” included the likes of IT procurement.
18 years later, the United Nations Development Program is almost unrecognizable, with Peiris overseeing the “change of direction of a massive tanker” to what was one of the “first agencies to have a digital office; one of the first to hire a Chief Digital Officer”, and the first to adopt a notion of “digital by default but not digital only”, a fundamental mechanism enabling UNDP to embed digital into the DNA of its program delivery across organizational levels.
Today, Peiris oversees many of these initiatives, working at the forefront of UNDP’s global digital transformation and data strategy. With its now advanced digital position, UNDP holds a unique place within the UN ecosystem, he begins our conversation by explaining. While many agencies have a specific mandate—such as the World Health Organization’s (WHO) focus on health - UNDP adopts a comprehensive approach that spans entire governments, the private sector, and societies through its “whole-of-society" approach. When a government from one of the 170+ countries where the UNDP operates seeks digital support, the organization is therefore able to connect the dots.
A systems-level approach to solving digital challenges
In practice, this means taking a systems-level approach to solving problems, “helping governments in their digital journey all the way from assessments – covering connectivity, regulation, business ecosystems, and people – to development and implementation”. This creates a coordinated approach between governments, civil society and the private sector. Together, they build ownership, support human-centered design, mitigate risks, and establish accountability throughout their digital transformation.
Crucial to this effort have been initiatives fostering the creation of digital-first mindsets across governments, particularly by encouraging data-driven decision-and policymaking. In a joint effort with the German Development Agency (GIZ), UNDP has for several years been running a “Data to Policy Navigator”, an initiative targeted at mid-senior level policymakers and designed to showcase examples of data-driven policy successes from a wide range of sectors and countries. In an interview at the UN World Data Forum 2023, Peiris highlighted the initiative’s unique accessibility – “most of the [existing] resources available target data scientists at National Bureaus of Statistics”, while with the navigator “no advanced or prior knowledge of data science is needed”.
The primary challenge in implementing such an initiative is largely a one of organizational culture, Peiris tells me a year on from that interview, one that comes back to the “random and disconnected” IT approaches of old. He highlights Moldova as an example of successful data-driven policy implementation. After being hit by a severe energy crisis in October 2021, the UNDP provided technical assistance to the country to create a “Fund for reduction of Energy Vulnerability”, developing an online platform for households to register to apply for financial compensation to reduce their energy bills.
Creating a decentralized needs-based compensation mechanism was challenging and marked the first platform to connect multiple sectors with the government. According to Peiris, the key to the success of the UNDP and Moldova collaboration was the “willingness to accept that their solution would not be bulletproof from the outset”, and to trust in non-traditional data sources to build the program. In contrast, countries that fail often treat digital initiatives as “graphs and data” without connecting them to real-world problems, and lack the cultural readiness to embrace failure while pushing the boundaries of digital innovation.
I ask Peiris how to address this reluctance , and what the UNDP can do to foster more initiatives like those in Moldova, Indonesia and Costa Rica which have successfully used geospatial and open-source data to tackle climate change and floods. “People often focus on the reluctants”, he tells me, “but working with those on the fence can naturally create the momentum for digital adoption”.
The UNDP, with boots on the ground in 170 countries, are able to demonstrate to those on the fence the success of initiatives in neighboring countries, leading study tours for decision-makers to other countries and highlighting best practices. Peiris emphasizes that the advantage of digital is their relatively quick impact: “unlike an education policy, which might take ten years to show results, digital transformation can sometimes be seen in just two to four years”.
Supporting countries in their AI Transformation
As in any sector, AI has been the hot topic in digital government for the past couple of years. The shift to AI-ready government will involve just a big a transformation as the ones that Peiris has driven within the UNDP and through the agency’s work. I ask him about the lessons from his previous experiences that could aid in managing this transition.
Transitioning the UNDP to a digital-ready organization involved embedding digital into the existing systems and processes of the organization, not merely treating it as an add-on. “The ideal point for this is at the project ideation phase”, he explains, describing an ERP system that introduced digital as a mandatory component for all UNDP projects. Every project “needed a digital indicator to ensure it was designed with a digital-first mindset”. Similarly, a transition to AI will involve integrating AI into current systems and processes while also creating new ones.
This journey, he notes, will be a gradual one toward maturity. He talks about the early days of the same ERP system in the UNDP, where digital indicators rarely stretched beyond “social media monitoring”. In the same way, governments should not try and walk before they can run with AI. “AI is an obvious candidate for change, but when you dig deeper they don’t have the fundamental things in place like data exchanges”.
To manage governments’ AI ambitions, and to support their digital transformation journey, the UNDP has designed an AI Readiness Assessment (AIRA) to equip policymakers with the insights on design and implementation that they need to progress on their AI journey. In a recent blog outlining the assessment, Peiris wrote that “countries are at different stages of their AI journey, and careful assessment is needed to determine the appropriate digital infrastructure, governance and enabling community” to deliver on their ambitions. The assessment will help countries better understand their current level of preparedness and what they need to adopt responsible, ethical and sustainable AI systems”.
Central to this effort is building local capacity for the AI revolution. “You can parachute in experts”, Peiris explains, “but you need someone local to ensure sustainable development and prevent governments from getting locked into vendor contracts they cannot escape ”. In the same way that a digital-first organization may start with 6,000 “monitoring social media” indicators, so too should governments’ AI revolutions start with a readiness assessment, or the AI Ethical Standards which the UNDP has been working on alongside UNESCO and AI as part of the United Nations Inter-Agency Working Group on AI. UNDP teams – with their own internal programs like Digital Fitness building their internal capacity – will no doubt continue to play a crucial role in supporting them.