World Solve

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.

Many apps still handle intermittent internet connectivity poorly, causing data loss or conflicts when switching between offline and online states. Truly seamless offline-first design remains inconsistently implemented across the software industry.

Attackers can sometimes convince phone carriers to transfer a victim's number to a new device, bypassing two-factor authentication tied to that number. Carriers have improved verification, but this attack method remains an ongoing security gap.

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.

Ensuring software works well with assistive technologies is often treated as a lower priority than testing for basic functionality or security, leading to inconsistent accessibility even in otherwise well-built products. Making accessibility testing equally rigorous by default remains unresolved.

Scammers increasingly use AI to generate convincing fake job postings and interview processes designed to steal personal information or money from desperate job seekers. Platforms are still catching up to detect and remove these scams effectively.

Some of the internet's most widely used open-source components are maintained by very few volunteers, and serious vulnerabilities have been found lurking undetected for years in code millions of systems depend on. Better resourcing and review processes for open-source security remain inconsistently funded.

When companies discontinue software or online services, users often lose access to their data or functionality with little warning or support for migrating elsewhere. Standardized, consumer-friendly transition processes remain largely absent from the industry.

Scammers increasingly use AI-cloned voices of family members to fake emergencies and pressure victims into sending money quickly, exploiting trust before people have time to verify. Reliable, fast verification methods to counter this remain underdeveloped.

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.

Insurers offering cybersecurity coverage often struggle to accurately assess a company's true risk level, leading to mispriced policies that either overcharge safe companies or undercharge vulnerable ones. Better, more accurate risk assessment tools remain underdeveloped.

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.

Low-code and no-code platforms have made software creation more accessible, but users without security expertise can still inadvertently build applications with serious vulnerabilities. Making these tools both accessible and inherently safe remains an unresolved design challenge.

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.

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.

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.

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.

Companies increasingly build products using AI models developed by others, but there's no consistent, scalable way to verify those models are safe and reliable before integration. This creates hidden risk throughout the growing AI supply chain.