World Solve

Future governments will need near-real-time visibility into commitments, arrears, off-budget liabilities, and contingent risks. The open problem is building financial transparency systems that ordinary citizens and auditors can actually use. This becomes particularly difficult in informal settlements, because public institutions must remain fair, resilient, and accountable while adapting to new risks.

A system can be factually sophisticated yet dangerous when users cannot distinguish knowledge, inference, speculation, and fabrication. The challenge is creating interfaces and evaluation standards that communicate uncertainty in ways people actually understand and use. The solution must work with limited capital, intermittent services, and local maintenance capacity. The research opportunity is to define measurable success criteria, test the approach in realistic settings, identify failure modes, and create a pathway from prototype to accountable deployment.

Many consultation processes collect comments but provide no traceable response or rationale. Structured argument maps, versioned policy records, and privacy-aware participation analytics could improve accountability without reducing deliberation to popularity. The opportunity is intentionally human-scale: a cooperative, clinic, workshop, school, or municipal team could pilot it within one locality and measure outcomes before expanding. Mathematical modeling can help with coordinate uncertainty, incentives, or resource constraints; entrepreneurship can turn a reliable workflow into a viable service.

Tutoring platforms can be targeted by individuals using AI-generated fake credentials to appear more qualified than they actually are, and verification systems don't always catch this convincingly fabricated background information. Closing this trust gap remains an unresolved platform challenge.

Purchased or AI-generated fake followers and engagement can make an account appear far more influential or trustworthy than it actually is, misleading both audiences and advertisers. Reliable detection and removal of this fake activity remains inconsistently enforced.

Online marketplaces can host sellers using fabricated business identities or locations, making it difficult for buyers to know who they're actually purchasing from. Reliable, scalable seller verification remains an unresolved marketplace trust issue.

Fake audio clips of political figures making statements they never actually said can spread quickly, and current detection tools don't always catch high-quality fakes before real damage to public understanding occurs. Improving detection speed and reliability remains urgent and unresolved.

Scammers can now use AI to generate convincing images of products that don't exist or don't match what's actually shipped, making it harder for shoppers to identify fraudulent listings. Detection tools on marketplaces remain inconsistently effective.

Cookie consent banners have become a near-universal nuisance without necessarily giving users meaningful, easy-to-understand control over their data. Redesigning this system to be both genuinely protective and user-friendly remains unresolved.

Recovering valuable materials from electronic waste is technically possible but often not profitable enough on its own to attract the investment needed to handle the sheer volume being generated. Making this economically viable at scale remains unresolved.

Without standardized measurement, it's difficult to compare the true environmental cost of different AI systems or hold companies accountable for improving efficiency. A consistent, transparent measurement standard remains undeveloped.

Terms of service and software licenses are often long, dense legal documents that almost no one actually reads before clicking agree. Making these terms genuinely understandable without losing their legal protection remains unresolved.

Search engines have improved at understanding context, but still frequently misinterpret nuanced or ambiguous queries, returning results that miss what the user actually wanted. True intent understanding remains an unresolved challenge in information retrieval.

Cybersecurity threats evolve quickly, but training programs for both professionals and everyday users often lag years behind the tactics attackers are actually using. Closing this education gap remains an ongoing structural challenge.

AI models often can't accurately tell when a question falls outside what they actually know, leading them to guess confidently rather than admit uncertainty. Teaching AI to recognize its own limits remains an unresolved research problem.

Surface meltwater can destabilise shelves in complex ways. Field experiments and high-fidelity models are needed.

Astronomical observations demand extra unseen mass, but candidate particles from extensions of the Standard Model remain undetected. New detector designs, astrophysical surveys, and theoretical models are needed to pin down its nature.

Big Bang models suggest equal amounts of matter and antimatter, yet the observable Universe is dominated by matter. Precision measurements of CP violation and new physics at high energies could explain the imbalance.

Observations show billion-solar-mass black holes early in cosmic history, which standard growth models struggle to explain. Improved surveys of high-redshift quasars and simulations of direct-collapse scenarios are needed.

Simulations predict substructure that observations only partly confirm. Strong lensing and stellar stream measurements can probe these scales.