Martin-dm/Getty
By Frank Swain
UK HEALTH secretary Jeremy Hunt has called on social media giants to do more to tackle sexting among the nation’s teens, which he blames for rising cases of mental illness. Yet his proposals for smart locks that stop teenagers sharing sexually explicit images are just the latest example of government demanding magical fixes for complex societal problems.
Giving evidence as part of a House of Commons inquiry into suicide prevention, Hunt singled out social media as a key platform for abuse, telling the panel, “I ask myself the simple question as to why you can’t prevent the texting of sexually explicit images by people under the age of 18.” He also asked why “word pattern recognition” couldn’t be used to identify and stop cyberbullying.
Hunt’s tech proposals are easy to suggest, but much harder to implement. Artificial intelligences can flag abusive keywords and recognise explicit images, but these are crude tools that often fail to understand context and are easy to circumvent.
Most practical filters, like those protecting the comments sections of a website, rely on feedback from thousands of viewers who can flag objectionable content, which isn’t much use in a two-way chat dialogue. And similar content filters on Facebook have resulted in women having their accounts suspended for sharing photos of breastfeeding.
How can we hope to build AI that recognises porn, when even US Supreme Court judges have failed to pin down what counts as obscene, only concluding, “I know it when I see it“? Even if we could, we shouldn’t. In the light of the pervasive powers granted ...
No comments:
Post a Comment