World Solve

Specialized AI hardware is improving, but there's still a significant tradeoff between processing power and energy efficiency that hasn't been fully solved. Breaking this tradeoff could dramatically reduce AI's growing environmental footprint.

Humans can learn a new task from a handful of examples, while most AI systems still need massive amounts of training data to perform reliably. Closing this efficiency gap remains one of AI research's central unsolved challenges.

The absence of robust, context-adapted solutions for education approaches that build the capacity of local institutions to manage climate risks creates cascading risks across interconnected human and natural systems. Key missing elements include durable performance, low lifetime cost, and governance arrangements capable of sustaining impact beyond pilot stages.

Citizens feel democracy is not delivering security or prosperity, driving support for radical solutions and coups. Civic-engagement platforms and institutional-performance dashboards can rebuild trust.

Regulatory and technical bottlenecks—including differing signalling systems and capacity constraints—slow modal shift. Integrated corridor-planning tools and interoperability solutions can boost rail usage.

Regulatory and technical bottlenecks—including differing signalling systems and capacity constraints—slow modal shift. Integrated corridor-planning tools and interoperability solutions can boost rail usage.

Existing tests focus on known-answer questions rather than open research tasks. Benchmarks that evaluate models on genuine research subtasks can guide development.

Magnetically confined plasmas exhibit instabilities and transport phenomena that limit performance. Better diagnostics and simulation tools can improve reactor design.

Buildings waste energy and materials. Green building standards can improve performance.

Standards vary, complicating comparison. Harmonized ESG metrics can help.

Urban private schools vastly outperform rural public schools in resources and outcomes, deepening long-term inequality. Digital learning platforms optimized for low-bandwidth rural connections could help level the field.

The absence of robust, context-adapted solutions for solutions for safe long-term storage and monitoring of captured carbon dioxide creates cascading risks across interconnected human and natural systems. Key missing elements include durable performance, low lifetime cost, and governance arrangements capable of sustaining impact beyond pilot stages.

Women perform the vast majority of line-level factory work in textile exports but are consistently passed over for supervisory, engineering, and managerial promotions.

Undiagnosed vision problems in children can quietly affect school performance for years before being caught. Early, affordable screening at scale remains unavailable in many school systems worldwide.

Some believe rewards should go strictly to those who work hardest or perform best, while others argue unequal starting points make pure merit-based systems inherently unfair. This tension remains philosophically unresolved.

Physical laws impose some theoretical limits on computation speed and energy use, but the exact universal boundary, if one truly exists, remains unresolved. This connects computer science directly to fundamental physics.

Some lower bounds on computational efficiency are proven for specific tasks, but a fully unified theory covering every possible algorithmic problem remains unresolved. This unresolved question shapes the outer limits of computer science.

Biodegradable packaging alternatives exist but often don't yet match the protective qualities of traditional plastic needed to keep food safe and fresh for the same duration. Closing this performance gap remains an unresolved materials science challenge.

Cameras and sensors used in security or autonomous systems still perform noticeably worse in poor visibility conditions than in daylight, creating real safety and reliability gaps. Closing this performance gap affordably remains unresolved.

Wireless charging remains noticeably slower and less energy-efficient than plugging in directly, limiting its adoption as a true primary charging method. Closing this performance gap remains an unresolved engineering challenge.