The Court of AI Capability
Can AI develop a chatbot that can empathetically support a person experiencing grief or loss?
⚖ Quase"The jury agreed that today’s models can mirror empathy but cannot truly feel or fully grasp grief, leaving them admirably close yet short of the mark."
Can AI create a personalized nutrition plan that takes into account a person's genetic profile, health goals, and dietary preferences?
⚖ Quase"With measured pragmatism, the jury granted the petition “almost,” recognizing that AI has mastered the mechanics of assembling genetic, health, and preference data into a tidy nutr…"
Can AI design a prosthetic limb that can be controlled by a person's thoughts and muscle signals?
⚖ Quase"The jury paused at the threshold of the possible, nodding at the dazzling demonstrations of mind-controlled prosthetics and the seamless chatter between neurons and machines, yet h…"
Can AI create a customized exercise plan that adapts to a person's physical limitations and goals over time?
⚖ Quase"The jury found that artificial intelligence has indeed learned to sketch out personalized exercise blueprints, but it still stumbles when trying to lace those blueprints with real-…"
Can AI create a new sport by combining elements of existing sports and defining its rules and objectives?
⚖ Quase"After lively deliberation, the jury acknowledged the generative spark of artificial creativity but found that spark insufficient to fully ignite an original sport; the system could…"
Can AI develop a system that can predict the success of a new product based on social media trends and consumer behavior?
⚖ Quase"The jury found the system promising yet premature, acknowledging AI’s keen eye for patterns in chatter and clicks but stopping short of trusting its crystal ball for success."
Can AI generate a realistic and engaging script for a podcast or radio show, including dialogue and sound effects?
⚖ Sim"The jurors found the task well within AI’s current toolbox, noting that large language models paired with audio-friendly formatting can spin out engaging scripts complete with dial…"
Can AI predict the likelihood of a social movement going viral based on its message and audience demographics?
⚖ Quase"The jury found AI capable of sketching the rough outlines of a viral social movement with far more accuracy than a coin flip, yet still stumbled when the crowd’s mood changed overnight."
Can AI convince a child to eat a vegetable they dislike?
⚖ Em análise"The jury grappled with the delicate chemistry of persuasion, and though no one claimed victory, the lone abstainer saw flickers of hope in chatbot charm."
Can AI make a moral judgment in a complex, real-world scenario?
⚖ Não"The jury found the question itself a bridge too far, concluding that moral judgments are not merely computations but acts of human conscience."
Can AI detect when a friend is on the edge?
⚖ Em análise"After spirited deliberation, the jury could not agree whether AI had crossed the threshold of emotional literacy, splitting between cautious optimism and skeptical refusal to certify."
Can AI know when to be quiet?
⚖ Em análise"The jury found itself tangled in the paradox of the question, unable to agree whether silence could ever be “known” by an algorithm when human ears still struggle to interpret it."
Can AI pick a christmas gift your mother will actually love?
⚖ Quase"The jury reached a near-unanimous stalemate, with only one juror leaning toward “almost” after determining that while an AI can mine preferences and suggest options with eerie prec…"
Can AI pass turn-based interaction turing tests for 5-minute windows?
⚖ Quase"After weighing testimony that today’s large language models can sail through short, tightly scripted Turing-style exchanges while still stumbling when the wind shifts or the topic …"
Can AI drive a car safely through complex urban environments?
⚖ Quase"After careful deliberation, the jury returned a near-unanimous verdict of ALMOST, finding that while autonomous vehicles can navigate controlled urban environments with impressive …"
Can AI detect counterfeit currency via image?
⚖ Sim"The jury swiftly sided with the affirmative, finding that AI’s sharp eye for detail makes it a capable sleuth against deceptive currency—provided the notes are clean and the lighting is right."
Can AI generate empathetic-sounding text replies?
⚖ Sim"The jury unanimously found that large language models can summon tones of empathy from their vast lexicons, rendering replies that often sound caring under carefully curated condit…"
Can AI win a go match against the world champion?
⚖ Sim"The jury returned a unanimous verdict in favor of today’s proposition, finding that artificial intelligence has indeed ascended to the summit of competitive Go."
Can AI translate asl video to english text in research conditions?
⚖ Quase"The jury concluded that ASL-to-English translation has made remarkable strides in controlled settings, where gestures are clear and vocabulary is limited, but still stumbles when c…"
Can AI generate working unit tests from a description of intent?
⚖ Sim"With a rare moment of unanimity, the jury found that current AI can reliably transform plain intent into working unit tests, citing Copilot and other code-savvy models as living proof."
Can AI generate court-ready first-draft contracts for common transactions?
⚖ Sim"The jury found no fault with the AI’s contract drafting, noting that it reliably converts plain-language descriptions into legally coherent first drafts without human prompting."
Can AI compose a fugue in the style of bach?
⚖ Quase"After careful consideration, the jury found that while artificial intelligence can mimic the intricate counterpoint and tonal architecture of a Bach fugue, it stumbles in sustainin…"
Can AI write a passable sonnet in the style of shakespeare?
⚖ Sim"The jury found, with unanimous delight, that while today’s models may not feel the heartache of a Stratford sonneteer, they can still craft fourteen iambic lines that rhyme and sca…"
Can AI generate a realistic deepfake video of a public figure speaking?
⚖ Sim"After hearing compelling evidence, the jury unanimously agreed that AI has reached the capability to generate realistic deepfake videos of public figures with remarkable fidelity."
But the data is real.