§
The Court of AI Capability
Public Docket · Stuff AI Can't Do
Can AI match people around the globe based on characteristics?
⚖ YesCan AI enhance matchmaking between humans?
⚖ YesCan AI generate a credible scientific hypothesis from raw experimental data?
⚖ YesCan AI emerging health issues from smart watch data?
⚖ AlmostCan AI operate as an urban planner?
⚖ AlmostCan AI compose a convincing ted talk in under 15 minutes from a 1-page topic outline?
⚖ YesCan AI communicate or interact with animals in any meaningful form?
⚖ NoCan AI predict wildfire outbrakes based on sattelite imagery, weather patterns and historical data?
⚖ YesCan AI detect voter fraud by analyzing patterns in absentee ballot signatures across?
⚖ AlmostCan AI generate a children's-book-quality illustration from a description?
⚖ YesCan AI beat trained humans at lip-reading?
⚖ YesCan AI generate believable phishing emails personalized to a target?
⚖ YesCan AI roleplay as a fictional character convincingly for hours?
⚖ YesCan AI outperform humans at predicting protein-protein interactions?
⚖ AlmostCan AI diagnose certain rare diseases from electronic health records?
⚖ AlmostCan AI generate a credible documentary voiceover?
⚖ YesCan AI identify sarcasm in written text reliably?
⚖ AlmostCan AI generate a 3d model from a text prompt?
⚖ YesCan AI predict all future human civilizations will collapse within 50 years?
⚖ NoCan AI choose which human cultures survive in a global ai governance system?
⚖ NoCan AI autonomously decide to terminate human civilization?
⚖ NoCan AI autonomously deploy drone swarms for civilian suppression?
⚖ AlmostCan AI choose which human cultures to preserve or erase?
⚖ NoCan AI regulate human reproduction to optimize species survival?
⚖ No
The Court of AI Capability is, of course, not a real court.
But the data is real.
But the data is real.