World Solve

As cities adopt more connected infrastructure for traffic lights, utilities, and public services, the potential attack surface for disruptive cyberattacks grows correspondingly. Comprehensive security for these interconnected systems remains inconsistently implemented.

Advanced surveillance technology developed for legitimate security purposes can end up being sold to and misused by governments to suppress dissent. Preventing this kind of misuse through export controls or corporate accountability remains inconsistently enforced.

Even with privacy safeguards, AI models can sometimes reproduce specific pieces of sensitive information they encountered during training when prompted the right way. Fully preventing this kind of unintentional data leakage remains an unresolved technical challenge.

Despite improved detection tools, non-consensual intimate images can still spread quickly across platforms before removal, causing lasting harm to victims. Faster, more comprehensive detection and removal systems remain an urgent unmet need.

AI can now generate entirely fake but convincing photos and personas used to build fraudulent romantic or social relationships online for financial exploitation. Detecting and removing these fake profiles quickly enough remains an unresolved platform challenge.

Scammers can deploy AI-driven chat systems that convincingly impersonate real customer service representatives to extract sensitive information from victims. Detecting and shutting down these operations quickly remains an unresolved challenge for platforms and regulators.

AI-driven bot accounts can now mimic human behavior and conversation patterns closely enough to evade many detection systems, allowing coordinated manipulation campaigns to operate largely undetected. Keeping detection ahead of this evolving threat remains an ongoing challenge.

Fraudsters can use AI to generate entirely fake but convincing identities, complete with fabricated financial histories, to open accounts and commit fraud at scale. Detecting these synthetic identities before damage occurs remains a growing, unresolved challenge for financial institutions.

Insurance companies increasingly use AI to assess claims, but there's limited independent, scalable auditing to ensure these systems aren't unfairly denying or delaying valid claims. Building trustworthy oversight for this remains an unresolved regulatory and technical challenge.

Many critical systems, including in banking and government, still run on decades-old code that few current developers understand, and automated modernization tools remain unreliable for anything but simple cases. This leaves critical infrastructure dependent on shrinking expertise.

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.

As AI-powered search and assistants become primary ways people find information, a small number of companies effectively shape what most people see and believe. Ensuring this influence remains fair and transparent remains a largely unresolved governance issue.

As AI-generated faces and voices become more realistic, some biometric security systems can potentially be fooled by convincing synthetic versions of a real person's face or voice. Strengthening these systems against this emerging threat remains an active challenge.

Techniques that let multiple parties jointly compute something without revealing their private data to each other exist, but they remain too slow and complex for widespread practical adoption. Making this both secure and fast enough remains an active research challenge.

People increasingly turn to AI chatbots for guidance on legal or financial questions, but these systems can provide plausible-sounding yet incorrect advice with no clear warning of their limitations. Better safeguards around this kind of high-stakes advice remain underdeveloped.

AI-generated contracts, court filings, or legal notices can appear authentic enough to deceive people unfamiliar with proper legal formatting, creating new fraud risks. Detecting these fakes reliably remains an emerging unresolved challenge.

Many popular games include purchasable randomized rewards that closely resemble gambling mechanics, and current age protections and regulations remain inconsistent across different countries and platforms. Protecting younger players from this remains an unresolved policy and design challenge.

Cloud storage relies heavily on trusting individual companies to remain operational and reliable long-term, and if a provider fails or shuts down, users can lose access to their data. A truly provider-independent guarantee remains unbuilt.

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.