World Solve

Utility data are fragmented, pressure changes are poorly localized, and repairs compete with emergency work. Low-cost sensing, probabilistic fault localization, and outcome-based maintenance could let local operators prioritize the few failures that matter most. 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 measure uncertainty, incentives, or resource constraints; entrepreneurship can turn a reliable workflow into a viable service.

Networks of AI-assisted fake dating profiles can be deployed at scale to build trust with vulnerable users before extracting money or personal information. Detecting these coordinated networks before real harm occurs remains an unresolved platform challenge.

Some financial institutions still use voice recognition as an authentication method, but increasingly realistic AI voice cloning threatens to undermine this security layer. Strengthening these systems against synthetic voice attacks remains an active, unresolved challenge.

AI chat systems can sometimes misread customer frustration or urgency, responding in ways that unintentionally worsen a situation rather than resolve it. Better emotional context understanding in these systems remains an unresolved design challenge.

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.

Bad actors can spread coordinated campaigns across several platforms at once, and current detection systems often work in isolation per platform rather than recognizing the broader cross-platform pattern. Building this kind of coordinated detection remains an unresolved industry challenge.

When ground-based internet infrastructure fails during disasters, many critical services lack an affordable, reliable satellite backup to maintain connectivity. Closing this resilience gap remains an unresolved infrastructure investment challenge.

Independent software developers and small studios can lose significant revenue to piracy networks that distribute their work without payment, and enforcement remains difficult and inconsistent across different countries. Protecting smaller creators specifically remains an underaddressed issue.

Existing assistive navigation tools for blind and visually impaired users remain limited in accuracy and are often too expensive for widespread daily use. A truly reliable, affordable solution remains an unmet accessibility need.

A huge share of global internet traffic between continents relies on a relatively small number of undersea cables, and damage to just one can disrupt connectivity for entire regions. Building enough redundancy to fully prevent this remains costly and incomplete.

Even after an AI model is retired, techniques exist that can sometimes extract fragments of its original training data, posing ongoing privacy risks. Fully preventing this kind of data leakage after decommissioning remains unresolved.

AI hiring systems trained on historical hiring data can inadvertently learn and repeat past patterns of discrimination, even when explicitly designed to avoid it. Fully eliminating this bias while maintaining useful screening capability remains unresolved.

While hardware energy use gets some attention, the carbon footprint of running specific pieces of software, from development to daily use, remains largely unmeasured and unaccounted for. Building this tracking into standard practice remains unresolved.

Automated trading systems can execute manipulative strategies at speeds far beyond human traders, and regulators still struggle to detect this kind of abuse in real time. Closing this detection gap remains an urgent unresolved challenge for financial oversight.

App permission systems exist, but many apps still collect more data than users realize or intend to allow, often through complex, hard-to-navigate settings. True granular control over personal data remains inconsistently implemented across platforms.

Content that provokes strong emotional reactions tends to get algorithmically boosted because it keeps people engaged longer, regardless of whether it's accurate or healthy. Redesigning these systems without destroying platform business models remains unresolved.

Most software vulnerabilities trace back to common coding mistakes that better language design could prevent from the start. Despite decades of research, mainstream programming still makes these mistakes easy to make.

mRNA and vector-based technologies can shorten response times, but manufacturing capacity and regulatory agility must be built in advance. Global manufacturing networks and adaptive trial designs will matter.

mRNA and vector-based technologies can shorten response times, but manufacturing capacity and regulatory agility must be built in advance. Global manufacturing networks and adaptive trial designs will matter.

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