A broadcaster is piloting a generative AI tool that auto-generates social media captions for live news clips. Select TWO risks the governance committee should treat as harmful-content concerns requiring mitigation before wider rollout. (Select TWO.)
Select all correct answers, then click Submit.
Short Explanation
A caption running long or looking slightly off-brand is just an annoyance for you to clean up later. A caption that mischaracterizes a tragedy or slips in biased language causes real harm to real people, and that's the line separating a formatting bug from a genuine harmful-content risk worth pausing your rollout for.
Full Explanation
Misrepresenting sensitive events and generating biased or offensive language are both genuine harmful-content risks: the first can cause real-world harm and reputational damage, and the second directly conflicts with editorial and responsible-AI standards, so both warrant mitigation before broader rollout. Running slightly over a platform's character limit is a formatting inconvenience that might require a quick fix, but it does not cause harm to viewers or the public. An inconsistent font style is a brand-consistency issue, not a harmful-content risk; inconsistent styling does not mislead or harm anyone. Taking an extra second to generate captions is a performance concern about latency, unrelated to the substance of what the captions say. The general principle: harmful-content risk assessment should focus on whether an output could mislead, offend, or cause real-world harm, not on unrelated formatting, branding, or performance issues that happen to occur in the same feature.