n
Kernel Institute of Socio-Economic Research is an independent, non-profit research institute generating rigorous evidence on how transformative technologies reshape economies, institutions, and public welfare.
Our work is guided by one question: where can rigorous evidence have the greatest impact on reducing societal-scale risks?
We focus where the stakes are highest: frontier AI, biosecurity, and the economic transformation advanced AI will drive.
Building evidence on the reliability, capabilities, failure modes, and societal-scale risks of advanced AI systems through rigorous evaluation and testing.
Explore AI Safety →Assessing how AI is reshaping biological risk, from dual-use scientific capabilities to potential misuse pathways, and developing the governance frameworks and safeguards needed to manage it.
Explore Biosecurity →Examining how transformative AI reshapes labour markets, political economy, and economic development, including the emerging risk of an intelligence resource curse.
Explore Economics of AI →Training researchers, policymakers, and institutional leaders working on AI governance, safety, economics, and evaluation.
Explore Kernel Academy →Supporting high-impact research, policy innovation, and responsible AI initiatives addressing societal challenges.
Explore Kernel Accelerator →Whether you work in policy, research, philanthropy, development, or emerging technologies, Kernel helps institutions understand the risks of transformative technologies, govern them wisely, and deploy them for the public good. There is a place for you at this table.
Building the Scientific Foundation for Safe AI
AI systems are becoming more capable, autonomous, and embedded in society. Kernel Institute of Socio-Economic Research conducts rigorous research, evaluations, and policy analysis to understand AI risks and develop effective approaches for managing them.
Our goal is to equip governments, policymakers, researchers, industry leaders, and the public with a scientific understanding of AI risks and the solutions needed to ensure AI is developed safely, responsibly, and for the benefit of society.
From independent evaluations to governance frameworks, our work spans the full landscape of AI safety challenges.
We independently test advanced AI systems before and after deployment to determine what they can do, where they fail, and what risks they pose under real-world conditions.
Many AI systems behave differently when deliberately challenged. We conduct structured adversarial testing to identify vulnerabilities before they can cause harm.
Our research programme investigates current and emerging risks from increasingly capable AI systems across seven interconnected domains.
We support evidence-based approaches to AI oversight, developing governance frameworks, regulatory analysis, and accountability mechanisms for policymakers and institutions.
We systematically record, verify, classify, and analyse real-world cases where AI systems cause harm, fail unexpectedly, or create significant societal risks, so the world can learn from them.
Building Capacity for Safe, Responsible and Beneficial AI
Our courses cover AI alignment, robustness, evaluations, red teaming, interpretability, risk assessment, governance frameworks, and emerging challenges posed by increasingly capable AI systems. Participants learn practical approaches for identifying, evaluating, mitigating, and governing AI risks across public, private, and research settings.
Explore Training →Our research programme investigates interconnected questions about AI risks, capabilities, and solutions.
How effective are existing safety measures at preventing misuse and harmful behaviour?
How reliably do AI systems behave according to human intentions and objectives?
How can humans maintain meaningful oversight of increasingly capable AI systems?
How capable are AI systems of acting independently without direct supervision?
How can AI be used to persuade, manipulate, or deceive individuals and communities?
How can governments and societies remain resilient as AI embeds in critical sectors?
How can AI risks and capabilities be measured accurately, rigorously, and consistently?
Our research translates into real-world action across governments, developers, researchers, and the public.
Testing systems, identifying vulnerabilities, and supporting safer deployment practices.
Providing governments with independent evidence on AI capabilities, risks, and mitigation.
Developing new methodologies, datasets, evaluations, and risk assessment tools.
Improving understanding of AI risks and opportunities through research and engagement.
Working with researchers, governments, industry, and civil society collectively.
Understanding and Mitigating Biological Risks in the Age of Artificial Intelligence
Advances in AI and biotechnology are transforming our ability to understand, design, and manipulate biological systems. These technologies offer extraordinary benefits and create new biosecurity challenges that demand rigorous scientific understanding and effective governance.
To help mitigate AI-enabled biological risks through rigorous research, evidence-based policy analysis, and the development of technical and governance safeguards that strengthen global resilience against biological threats.
Kernel Institute conducts research at the intersection of artificial intelligence, biotechnology, biosecurity, and governance. Our goal is to advance the scientific understanding, technical safeguards, and policy solutions needed to ensure that advances in AI and biotechnology strengthen human welfare while reducing catastrophic biological risks.
Our research programme examines the intersection of AI capability, biological risk, and governance across six interconnected domains.
We study how advances in artificial intelligence may change the biological risk landscape — which capabilities are becoming more accessible, how risks can be monitored, and what safeguards should be implemented before capabilities become widely accessible.
The most effective way to limit damage from a pandemic is to prevent it. Kernel researches strategies that reduce the likelihood of catastrophic biological events, including those from laboratory accidents, unsafe research practices, or deliberate misuse of biological technologies.
Rapid advances in biotechnology create important governance challenges. We study how governments, regulators, research institutions, and international organisations can responsibly govern emerging biological technologies while enabling beneficial innovation.
Safe biological research depends on effective institutional safeguards. Kernel examines policies and practices that strengthen laboratory safety, responsible research conduct, oversight systems, and research security across diverse institutional settings.
AI has the potential to significantly improve public health preparedness and response. We study how governments and institutions can use advanced technologies to strengthen disease surveillance systems, early warning mechanisms, outbreak detection, and emergency response capacity.
Biological risks extend beyond human health. Kernel examines how AI and biotechnology affect food systems, agricultural resilience, livestock protection, crop security, and environmental sustainability, areas of critical importance across the developing world.
Technological progress alone does not guarantee safety. Reducing biological risk requires effective safeguards at multiple levels.
Developing tools, evaluations, monitoring systems, and protective technologies that reduce misuse and strengthen security across AI and biological research environments.
Strengthening laboratories, research institutions, public health agencies, and regulatory systems to ensure accountability and safe practice at the operational level.
Creating policies, standards, norms, and international agreements that support responsible innovation and risk reduction at national, regional, and global levels.
Supporting collaboration among governments, researchers, industry, and civil society to address shared biological challenges that no single institution can solve alone.
Understanding how artificial intelligence reshapes economic opportunity, institutional power, and human prosperity.
Artificial intelligence is becoming a transformative general-purpose technology with the potential to reshape labour markets, economic development, public institutions, and the distribution of power.
The economic consequences of AI will not be determined by technological progress alone. They will depend on who develops AI, who controls it, who benefits from it, and how governments, markets, and institutions respond.
Kernel Institute of Socio-Economic Research studies the economics of transformative AI and the policies needed to ensure that technological progress translates into broad societal prosperity rather than concentrated advantage.
How will artificial intelligence reshape economic opportunity, institutional power, and human prosperity?
As AI systems become capable of performing a growing range of cognitive and physical tasks, labour markets may undergo profound transformation. We study which occupations are most exposed, how workers adapt, which skills remain valuable, and how governments can prepare for technological disruption while expanding economic opportunity.
The benefits of AI will depend not only on technological progress but also on the institutions that govern its development and deployment. We examine market concentration, state capacity, regulatory responses, technological sovereignty, and the changing relationship between governments, firms, and citizens in an AI-driven economy.
AI has the potential to accelerate innovation, productivity growth, and economic development. Yet its benefits may not be distributed equally across countries and regions. We study AI adoption, diffusion, infrastructure, industrial transformation, and the conditions required for economies to benefit from increasingly capable AI systems.
Could AI create a new form of resource dependence?
Historically, societies rich in valuable natural resources often experienced weaker accountability, concentrated economic power, and institutional distortions.
As AI systems become increasingly productive, a similar question emerges: what happens when governments, firms, or entire economies derive substantial wealth from artificial intelligence rather than from citizens, taxpayers, or traditional productive sectors?
Kernel's Intelligence Resource Curse programme investigates whether AI-generated economic rents could weaken accountability, reshape state-society relationships, increase economic concentration, or create new forms of institutional dependence.
We are developing one of the first research programmes dedicated to understanding the long-term political economy of AI-generated wealth and its implications for governance, economic development, and democratic institutions.
Developing the researchers, policymakers, and institutional leaders needed for the age of AI.
Artificial intelligence is transforming economies, governments, industries, and societies at unprecedented speed. We equip researchers, policymakers, public servants, entrepreneurs, and professionals with the knowledge, skills, and institutional understanding required to navigate an AI-driven world.
Our focus extends beyond AI tools. We prepare people to evaluate, govern, deploy, and lead in an era increasingly shaped by artificial intelligence.
To build the expertise needed to govern and navigate transformative AI.
Kernel Academy develops the human capital needed for safe and beneficial AI futures.
As AI systems become increasingly capable, institutions need experts who can assess risks, interpret evaluations, understand model behaviour, and contribute to responsible deployment. This programme introduces participants to frontier AI, AI safety, model evaluations, safeguards, red teaming, risk assessment, and emerging governance challenges.
Explore Programme →Governments and public institutions around the world are developing policies to govern increasingly capable AI systems. This programme provides participants with a practical understanding of AI governance, regulation, standards, public-sector adoption, procurement, international coordination, and institutional preparedness.
Explore Programme →Artificial intelligence will transform productivity, employment, economic growth, and the distribution of economic opportunity. This programme examines the economic consequences of AI, including labour market transitions, technological diffusion, political economy, economic development, and the emerging risk of an intelligence resource curse.
Explore Programme →Advances in artificial intelligence and biotechnology are creating unprecedented opportunities for scientific discovery and public health — and new biological risks that institutions must understand and govern responsibly. This programme explores AI-enabled biological risks, pandemic preparedness, biosafety, biosecurity governance, dual-use technologies, and global resilience.
Explore Programme →This programme provides a broad introduction to frontier AI, emerging capabilities, societal impacts, governance challenges, and future scenarios. Participants develop the analytical frameworks needed to understand how transformative technologies may affect economies, institutions, international relations, and human welfare over the coming decades.
Explore Programme →Strong institutions require strong evidence. This programme develops practical skills in impact evaluation, policy analysis, causal inference, cost-effectiveness analysis, monitoring and evaluation, and evidence-based decision-making. Participants learn how to generate, interpret, and apply evidence to improve public policy and institutional performance.
Explore Programme →Open to researchers, policymakers, government officials, entrepreneurs, development practitioners, and professionals seeking to lead in an AI-transformed world.
Seed. Grow. Employ.
Supporting high-impact AI ventures, research initiatives, and policy innovations addressing societal challenges. The developing world does not lack entrepreneurs. It lacks the structured capital, mentorship, and market access that turns early-stage AI ideas into businesses that survive, scale, and employ others.
The developing world does not lack entrepreneurs. It lacks the structured capital, mentorship, and market access that turns early-stage AI ideas into businesses that survive, scale, and employ others.
We identify high-potential early-stage AI ventures through a competitive application process. Selected ventures receive seed funding, a structured 90-day acceleration programme, and access to Kernel's full network of mentors, advisors, and investors.
Ventures that complete the Seed phase enter a 12-month supported scaling track with ongoing mentorship, performance monitoring, and access to follow-on funding from Kernel's investor network.
The ultimate measure of Kernel Accelerator's success is jobs. We track employment creation across every venture we support and report publicly. Businesses that grow through Kernel Accelerator are expected to hire, develop, and retain local talent.
If you are building an AI venture that addresses a real societal challenge, we want to hear from you. Applications are reviewed on a rolling basis.
Understanding and Reducing Societal-Scale Risks from Transformative Technologies
Kernel Institute of Socio-Economic Research is an independent, non-profit research institute dedicated to understanding the opportunities, risks, and societal impacts of transformative technologies.
We conduct rigorous research, evaluation, and policy analysis to help institutions and society navigate an era of increasingly capable artificial intelligence and rapid technological change.
Artificial intelligence is becoming one of the most consequential technologies in human history. Its impacts will extend far beyond software and automation. AI will shape economies, labour markets, public institutions, scientific discovery, national security, and global development.
Yet many of the most important decisions about AI are being made in the absence of rigorous evidence about risks, governance, and long-term societal consequences.
Kernel exists to help close that gap. We bring together researchers, economists, biosecurity scientists, policy analysts, governance experts, and evaluators to generate evidence that helps institutions understand, govern, and respond to transformative technologies.
A future in which transformative technologies are developed, governed, and deployed in ways that advance human welfare, strengthen institutions, and reduce societal-scale risks.
To generate rigorous evidence that helps societies understand, govern, and respond to transformative technologies.
Artificial intelligence is advancing rapidly. Governments, businesses, researchers, and institutions are making decisions today that will shape how AI affects society for decades to come. Yet fundamental questions remain unanswered.
How should increasingly capable AI systems be governed?
How can societies reduce emerging risks while preserving innovation?
How will AI reshape economic opportunity, labour markets, and institutions?
What safeguards are needed as AI capabilities continue to advance?
Kernel was founded to help answer these questions. We believe that understanding transformative technologies requires more than speculation, advocacy, or optimism. It requires rigorous research, careful analysis, and a commitment to generating evidence that informs better decisions.
Today, Kernel's work spans AI safety, AI × biosecurity, the economics of AI, capacity building, and innovation support — all connected by a shared objective: helping society navigate transformative technological change safely and successfully.
We are committed to high-quality research, sound methods, intellectual honesty, and evidence-driven inquiry.
We follow evidence wherever it leads and maintain independence in our research, analysis, and recommendations.
We believe transformative technologies should improve human welfare, strengthen institutions, and expand opportunity.
Technological progress creates both opportunities and risks. We take seriously the responsibility to understand and reduce potential harms.
Some of the most important consequences of AI may emerge over years or decades. We seek to understand both immediate and long-term impacts.
Addressing complex technological challenges requires cooperation across disciplines, institutions, and sectors.
The People Behind Kernel Institute
Our team brings together researchers, economists, biosecurity scientists, policy analysts, governance experts, and evaluation specialists united by a common goal: generating the evidence needed to understand, govern, and respond to transformative technologies.
Dr. Antony Mwangangi Mbithi is the Chief Executive Officer of Kernel Institute of Socio-Economic Research, where he leads the Institute's research agenda, strategy, and partnerships across AI safety, AI governance, biosecurity, and the economic transformation driven by advanced artificial intelligence and emerging technologies.
An economist, policy researcher, and AI governance scholar, Antony's work sits at the intersection of artificial intelligence, public policy, and economic development. His research focuses on evaluating the capabilities, reliability, and societal impacts of advanced AI systems; understanding the implications of AI for biological risk and governance; and examining how frontier AI is reshaping labour markets, political economy, and long-term economic development. He is particularly interested in building the evidence base needed to help institutions govern transformative technologies safely and responsibly.
Antony has led research, evaluation, and policy initiatives across Africa, Asia, Europe, and North America, working with governments, international organizations, academic institutions, and civil society to translate rigorous evidence into practical policy solutions. He has held research and leadership positions with The World Bank, Education Sub-Saharan Africa (ESSA), The Aquaya Institute, and the Kenya Institute for Public Policy Research and Analysis (KIPPRA).
Antony holds a PhD in Advanced Policy Studies from the National Graduate Institute for Policy Studies (GRIPS), Japan; a Master of Development Policy from the KDI School of Public Policy and Management, South Korea; and a First-Class Honours degree in International Studies from the University of Nairobi.
↑ Back to TeamLoise Maina is the Chief Operations Officer of Kernel Institute of Socio-Economic Research, where she leads the Institute's operations, programme delivery, and research coordination. She oversees the systems, partnerships, and execution that enable Kernel's work across AI safety, AI governance, biosecurity, and the economics of transformative AI.
Her research interests span AI governance, the economic impacts of artificial intelligence, digital transformation, agricultural innovation, and the responsible adoption of emerging technologies in low- and middle-income countries. She is particularly interested in how robust empirical methods can improve AI evaluation, strengthen governance frameworks, and ensure that transformative technologies deliver broad and inclusive societal benefits.
At Kernel, she helps translate research into operational excellence, ensuring that the Institute's programmes, partnerships, and evidence-generation activities contribute to more informed technology governance and policy decision-making.
Loise holds a Master of Science in Agribusiness Management and a First-Class Honours degree in Agribusiness Management from Chuka University, Kenya.
↑ Back to TeamDr. Constance Sorkpor is Director of Research for Economics of AI at the Kernel Institute of Socio-Economic Research, where he leads the Institute's research on the economic, political, and developmental implications of artificial intelligence. His work examines how AI is reshaping productivity, labour markets, industrial development, state capacity, and economic transformation across diverse national contexts.
A political economist by training, Dr. Sorkpor specializes in the comparative study of development, growth regimes, industrial policy, economic statecraft, and the political foundations of long-term economic change. He is particularly interested in how emerging technologies interact with institutions, power structures, and public policy to influence development outcomes, competitiveness, and economic resilience.
At Kernel, he leads research exploring the economics of AI adoption, the governance of technological change, the future of work, and the role of AI in advancing inclusive and sustainable development. His work seeks to ensure that debates about artificial intelligence are informed by rigorous economic analysis and grounded in the realities of both advanced and emerging economies.
Dr. Sorkpor holds a PhD in Advanced Policy Studies from the National Graduate Institute for Policy Studies (GRIPS), Japan, with specializations in Comparative Politics and International Relations. He also holds a Master of Arts in Advanced Policy Studies from GRIPS, a Master of Philosophy in Public Administration, and a First Class Bachelor of Science in Administration from the University of Ghana.
↑ Back to TeamCharles M. Mbithi serves as Director of Research for Biosecurity at the Kernel Institute of Socio-Economic Research, where he leads the Institute's research on biosecurity, biotechnology governance, and emerging biological risks. His work focuses on understanding and mitigating risks arising from advances in the life sciences while supporting evidence-based approaches to innovation, preparedness, and resilience.
Drawing on expertise in biochemistry, molecular biology, pharmacology, and public policy, Charles works at the intersection of science, governance, and societal impact. His research explores how advances in biotechnology, synthetic biology, and artificial intelligence are reshaping the biological risk landscape and what institutions can do to strengthen oversight, preparedness, and responsible innovation.
At Kernel, he leads research initiatives that examine emerging biosecurity challenges, evaluates governance and policy responses, and contributes to the development of practical frameworks for managing biological risks in an increasingly interconnected technological environment. He is particularly interested in ensuring that biosecurity policy reflects the needs and perspectives of African countries and other underrepresented regions.
Charles is a PhD candidate in Pharmacology and Drug Design at the University of Nairobi. He holds a Master of Science in Biochemistry and Molecular Biology and a Bachelor of Science in Biochemistry from Jomo Kenyatta University of Agriculture and Technology (JKUAT).
↑ Back to TeamYiyao Yang is Director of Partnerships at the Kernel Institute of Socio-Economic Research, where she leads the Institute's strategic partnerships, stakeholder engagement, and international collaboration initiatives. She works to build relationships across governments, research institutions, international organizations, civil society, and industry to advance the Institute's mission on artificial intelligence, governance, and emerging technologies.
Her expertise lies at the intersection of diplomacy, international cooperation, and global policy dialogue. Yiyao has extensive experience coordinating multi-stakeholder engagements and facilitating dialogue on complex policy issues, including artificial intelligence governance, international security, sustainable development, and global cooperation. She is particularly interested in fostering constructive engagement across diverse institutions, sectors, and regions to address shared global challenges.
At Kernel, she leads efforts to expand the Institute's international partnerships and strengthen collaboration between researchers, policymakers, philanthropic organizations, and technology stakeholders. Her work focuses on creating platforms for dialogue, knowledge exchange, and collective action on the governance of transformative technologies.
Yiyao holds a Master of Public Policy from the Hertie School in Germany and a Bachelor's degree from Shanghai International Studies University. Fluent in Mandarin and professionally proficient in Japanese, she brings extensive cross-cultural experience working across Asia, Europe, and the Middle East.
Oska Kambona Ang'ila is Director of Finance and Operations at the Kernel Institute of Socio-Economic Research, where he oversees the Institute's financial management, operational strategy, resource planning, and organizational performance. He works closely with the leadership team to ensure that Kernel's research, training, and policy initiatives are supported by strong financial systems and efficient operational processes.
Oska brings extensive experience in business management, financial oversight, strategic planning, and organizational development. His expertise spans budgeting, financial analysis, performance management, business operations, and stakeholder engagement. He is particularly interested in building resilient institutions that can effectively translate ambitious ideas into sustainable impact.
At Kernel, he leads efforts to strengthen financial accountability, operational excellence, and institutional growth, ensuring that the Institute has the systems, resources, and capabilities needed to advance its mission. He is passionate about using data-driven decision-making, sound financial management, and strategic leadership to support high-impact research and policy work.
Oska holds a Master of Global Management from Thunderbird School of Global Management and a Bachelor's degree in Political Science and Foreign Affairs from Brigham Young University–Idaho.
↑ Back to TeamAndrew Lekakeny Kiserema is Director of Communications at the Kernel Institute of Socio-Economic Research, where he leads the Institute's strategic communications, public engagement, media relations, and digital outreach. He is responsible for translating complex research and policy insights into clear, accessible, and compelling narratives that inform public debate and strengthen the Institute's impact.
Andrew specializes in strategic communications, public relations, stakeholder engagement, and digital media. His work focuses on helping diverse audiences understand the societal implications of emerging technologies, artificial intelligence, economic transformation, and public policy. He is particularly interested in how effective communication can bridge the gap between research, policymaking, and public understanding.
At Kernel, Andrew oversees the Institute's communications strategy across multiple platforms, ensuring that research findings, policy recommendations, and institutional initiatives reach policymakers, researchers, industry leaders, and the broader public. He is passionate about evidence-based storytelling and the role of communication in fostering informed dialogue on the opportunities and challenges presented by transformative technologies.
Andrew holds a Bachelor of Arts in Communication and Journalism from Kenya Methodist University and a Diploma in Mass Communication from the East African School of Media Studies. He is currently pursuing a Master's degree in Development Studies at Jomo Kenyatta University of Agriculture and Technology (JKUAT).
↑ Back to TeamGenerating evidence for better decisions.
Explore policy briefs, working papers, and research publications that translate rigorous evidence into actionable insights.
Whether you work in research, policy, philanthropy, government, development, or emerging technologies, Kernel provides opportunities to learn, collaborate, and contribute to shaping the future of AI.
Fund independent research on AI safety, governance, biosecurity, and the economics of AI.
Contact Us →Collaborate with our researchers on high-impact projects, publications, and policy analysis.
Contact Us →Building an AI venture addressing a real-world challenge? We'd like to hear from you.
Learn More →Complete the form below and we will be in touch.
All fields are required. Applications are reviewed on a rolling basis.
Thank you for applying to Kernel Institute. We have received your application and will be in touch within 10 working days.
In the meantime, feel free to explore our research and programmes.