Verified accountThis account is verified. Learn more
@voxeu
VoxEU is a policy portal run by @cepr_org. It posts research based policy analysis from leading economists. Visit our development economics branch @vox_dev
For three decades, software companies have paid independent researchers to identify security vulnerabilities in their code. But how effective are bug bounty programmes, and how is AI changing the way software flaws are discovered?
In the latest episode of VoxTalks Economics, Tim Phillips speaks with @tkretschmer_LMU (@ImperialBiz and CEPR) about his research into the design and effectiveness of bug bounty programmes.
Drawing on data from the Bugcrowd platform, they discuss the trade-offs between open and invitation-only programmes. Opening programmes to more researchers can increase the number of vulnerabilities discovered, but also brings more duplicate and lower-quality submissions, increasing the burden of reviewing them.
They also consider what AI-assisted vulnerability detection means for the future of bug bounties, and why evaluating submissions remains a crucial part of software security.
In the fourth episode of the Economic Policy Series on #VoxTalks Economics, Tim Phillips is joined by @alejandrowerne7 (@GUAmericas) to discuss new research with Nicol谩s Magud and Antonio Spilimbergo examining more than 50 years of populist governments in Latin America.
The conversation looks at what happens after populist leaders take office. Comparing populist and non-populist governments, the research finds differences across a range of economic and institutional outcomes, including government expenditure and fiscal deficits, institutional strength, international reserves and, over time, inflation.
They also discuss why Latin America鈥檚 most recent wave of populist governments was able to remain in power for longer than many of its predecessors, including the role played by favourable commodity prices.