A recent user tested Google's AI by asking, "How much screen time is too much for teenagers?" Rather than providing links, as Google typically does, the AI gave a detailed answer that discussed the complexity of screen time. It noted that the quality and balance of time might matter more than the number of hours. The AI also said that "too much" time could depend on factors like a teenager’s sleep, exercise, school demands, and mood. The user then asked, "Should I take a daily aspirin?" This time, the AI provided medical information, warned about risks, and offered more tailored guidance if the user provided their age and medical history. These responses were described as authoritative and useful, though they differed in approach. The AI's responses can be categorized into four types: factual, interpretive, constructive, and strategic. A factual claim can often be checked against a source. An interpretive reply is built on evidence but may have multiple interpretations. A constructive answer is created rather than discovered, such as drafting a cover letter or writing a eulogy. A strategic answer combines information with judgment about goals, risks, trade-offs, and personal circumstances. For the screen time question, the AI initially mentioned a two-hour limit for teenagers but then noted that pediatric guidance emphasizes the quality and context of screen use over simple hours. The American Academy of Pediatrics states there is no exact recommended amount for teens and emphasizes the kind of screen use and what activities it might be displacing. For the aspirin question, the AI warned about risks, advised consulting a medical professional, and offered more tailored information if the user provided additional details. This caution aligns with the U.S. Preventive Services Task Force guidance, which recommends individualized decisions based on cardiovascular benefit and bleeding risk. Users are encouraged to recognize the type of answer they receive and consider whether the interpretation is persuasive or the advice fits their circumstances. Before asking whether an AI answer is right, users should ask what kind of answer it is, as the type will determine the next steps.