How tax data unlocks new insights for industrial policy


OECD article: “Value-added tax (VAT) is a consumption tax applied at each stage of the supply chain whenever value is added to goods or services. Businesses collect and remit VAT. The VAT data that are collected represent a breakthrough in studying production networks because they capture actual transactions between firms at an unprecedented level of detail. Unlike traditional business surveys or administrative data that might tell us about a firm’s size or industry, VAT records show us who does business with whom and for how much.

This data is particularly valuable because of its comprehensive coverage. In Estonia, for example, all VAT-registered businesses must report transactions above €1,000 per month, creating an almost complete picture of significant business relationships in the economy.

At least 15 countries now have such data available, including Belgium, Chile, Costa Rica, Estonia, and Italy. This growing availability creates opportunities for cross-country comparison and broader economic insights…(More)”.

Governing in the Age of AI: Building Britain’s National Data Library


Report by the Tony Blair Institute for Global Change: “The United Kingdom should lead the world in artificial-intelligence-driven innovation, research and data-enabled public services. It has the data, the institutions and the expertise to set the global standard. But without the right infrastructure, these advantages are being wasted.

The UK’s data infrastructure, like that of every nation, is built around outdated assumptions about how data create value. It is fragmented and unfit for purpose. Public-sector data are locked in silos, access is slow and inconsistent, and there is no system to connect and use these data effectively, or any framework for deciding what additional data would be most valuable to collect given AI’s capabilities.

As a result, research is stalled, AI adoption is held back, and the government struggles to plan services, target support and respond to emerging challenges. This affects everything from developing new treatments to improving transport, tackling crime and ensuring economic policies help those who need them. While some countries are making progress in treating existing data as strategic assets, none have truly reimagined data infrastructure for an AI-enabled future…(More)”

The Preventative Shift: How can we embed prevention and achieve long term missions


Paper by Demos (UK): “Over the past two years Demos has been making the case for a fundamental shift in the purpose of government away from firefighting in public services towards preventing problems arriving. First, we set out the case for The Preventative State, to rebuild local, social and civic foundations; then, jointly with The Health Foundation, we made the case to change treasury rules to ringfence funding for prevention. By differentiating between everyday spending, and preventative spending, the government could measure what really matters.

There has been widespread support for this – but also concerns about both the feasibility of measuring preventative spending accurately and appropriately but also that ring-fencing alone may not lead to the desired improvements in outcomes and value for money.

In response we have developed two practical approaches, covered in two papers:

  • Our first paper, Counting What Matters, explores the challenge of measurement and makes a series of recommendations, including the passage of a “Public Investment Act”, to show how this could be appropriately achieved.
  • This second paper, The Preventative Shift, looks at how to shift the culture of public bodies to think ‘prevention first’ and target spending at activities which promise value for money and improve outcomes…(More)”.

Nonprofits, Stop Doing Needs Assessments.


Design for Social Impact: “Too many non-profits and funders still roll into communities with a clipboard and a mission to document everything “missing.”

Needs assessments have become a default tool for diagnosing deficits, reinforcing a saviour mentality where outsiders decide what’s broken and needs fixing.

I’ve sat in meetings where non-profits present lists of what communities lack:

  • “Youth don’t have leadership skills”
  • “Parents don’t value education”
  • “Grassroots organisations don’t have capacity”

The subtext? “They need us.”

And because funding is tied to these narratives of scarcity, organisations learn to describe themselves in the language of need rather than strength—because that’s what gets funded…Now, I’m not saying that organisations or funders should never ask people what their needs are. The key issue is how needs assessments are framed and used. Too often, they use extractive “data” collection methodologies and reinforce top-down, deficit-based narratives, where communities are defined primarily by what they lack rather than what they bring.

Starting with what’s already working (asset mapping) and then identifying what’s needed to strengthen and expand those assets is different from leading with gaps, which can frame communities as passive recipients rather than active problem-solvers.

Arguably, a balanced synergy between assessing needs and asset mapping can be powerful—so long as the process centres on community agency, self-determination, and long-term sustainability rather than diagnosing problems for external intervention.

Also, asset-based mapping to me does not mean that you swoop in with the same clipboard and demand people document their strengths…(More)”.

The Missing Pieces in India’s AI Puzzle: Talent, Data, and R&D


Article by Anirudh Suri: “This paper explores the question of whether India specifically will be able to compete and lead in AI or whether it will remain relegated to a minor role in this global competition. The paper argues that if India is to meet its larger stated ambition of becoming a global leader in AI, it will need to fill significant gaps in at least three areas urgently: talent, data, and research. Putting these three missing pieces in place can help position India extremely well to compete in the global AI race.

India’s national AI mission (NAIM), also known as the IndiaAI Mission, was launched in 2024 and rightly notes that success in the AI race requires multiple pieces of the AI puzzle to be in place.3 Accordingly, it has laid out a plan across seven elements of the “AI stack”: computing/AI infrastructure, data, talent, research and development (R&D), capital, algorithms, and applications.4

However, the focus thus far has practically been on only two elements: ensuring the availability of AI-focused hardware/compute and, to some extent, building Indic language models. India has not paid enough attention to, acted toward, and put significant resources behind three other key enabling elements of AI competitiveness, namely data, talent, and R&D…(More)”.

How Innovation Ecosystems Foster Citizen Participation Using Emerging Technologies in Portugal, Spain and the Netherlands


OECD Report: “This report examines how actors in Portugal, Spain and the Netherlands interact and work together to contribute to the development of emerging technologies for citizen participation. Through in-depth research and analysis of actors’ motivations, experiences, challenges, and enablers in this nascent but promising field, this paper presents a unique cross-national perspective on innovation ecosystems for citizen participation using emerging technology. It includes lessons and concrete proposals for policymakers, innovators, and researchers seeking to develop technology-based citizen participation initiatives…(More)”.

Advanced Flood Hub features for aid organizations and govern


Announcement by Alex Diaz: “Floods continue to devastate communities worldwide, and many are pursuing advancements in AI-driven flood forecasting, enabling faster, more efficient detection and response. Over the past few years, Google Research has focused on harnessing AI modeling and satellite imagery to dramatically accelerate the reliability of flood forecasting — while working with partners to expand coverage for people in vulnerable communities around the world.

Today, we’re rolling out new advanced features in Flood Hub designed to allow experts to understand flood risk in a given region via inundation history maps, and to understand how a given flood forecast on Flood Hub might propagate throughout a river basin. With the inundation history maps, Flood Hub expert users can view flood risk areas in high resolution over the map regardless of a current flood event. This is useful for cases where our flood forecasting does not include real time inundation maps or for pre-planning of humanitarian work. You can find more explanations about the inundation history maps and more in the Flood Hub Help Center…(More)”.

Policymaking assessment framework


Guide by the Susan McKinnon Foundation: “This assessment tool supports the measurement of the quality of policymaking processes – both existing and planned – across  sectors. It provides a flexible framework for rating public policy processes using information available in the public domain. The framework’s objective is to simplify the path towards best practice, evidence-informed policy.

It is intended to accommodate the complexity of policymaking processes and reflect the realities and context within which policymaking is undertaken. The criteria can be tailored for different policy problems and policy types and applied across sectors and levels of government.

The framework is structured around five key domains:

  1. understanding the problem
  2. engagement with stakeholders and partners
  3. outcomes focus
  4. evidence for the solution, and
  5. design and communication…(More)”.

Intellectual property issues in artificial intelligence trained on scraped data


OECD Report: “Recent technological advances in artificial intelligence (AI), especially the rise of generative AI, have raised questions regarding the intellectual property (IP) landscape. As the demand for AI training data surges, certain data collection methods give rise to concerns about the protection of IP and other rights. This report provides an overview of key issues at the intersection of AI and some IP rights. It aims to facilitate a greater understanding of data scraping — a primary method for obtaining AI training data needed to develop many large language models. It analyses data scraping techniques, identifies key stakeholders, and worldwide legal and regulatory responses. Finally, it offers preliminary considerations and potential policy approaches to help guide policymakers in navigating these issues, ensuring that AI’s innovative potential is unleashed while protecting IP and other rights…(More)”.

Building AI for the pluralistic society


Paper by Aida Davani and Vinodkumar Prabhakaran: “Modern artificial intelligence (AI) systems rely on input from people. Human feedback helps train models to perform useful tasks, guides them toward safe and responsible behavior, and is used to assess their performance. While hailing the recent AI advancements, we should also ask: which humans are we actually talking about? For AI to be most beneficial, it should reflect and respect the diverse tapestry of values, beliefs, and perspectives present in the pluralistic world in which we live, not just a single “average” or majority viewpoint. Diversity in perspectives is especially relevant when AI systems perform subjective tasks, such as deciding whether a response will be perceived as helpful, offensive, or unsafe. For instance, what one value system deems as offensive may be perfectly acceptable within another set of values.

Since divergence in perspectives often aligns with socio-cultural and demographic lines, preferentially capturing certain groups’ perspectives over others in data may result in disparities in how well AI systems serve different social groups. For instance, we previously demonstrated that simply taking a majority vote from human annotations may obfuscate valid divergence in perspectives across social groups, inadvertently marginalizing minority perspectives, and consequently performing less reliably for groups marginalized in the data. How AI systems should deal with such diversity in perspectives depends on the context in which they are used. However, current models lack a systematic way to recognize and handle such contexts.

With this in mind, here we describe our ongoing efforts in pursuit of capturing diverse perspectives and building AI for the pluralistic society in which we live… (More)”.