5 Ways AI Can F*ck Up the US Election
Don’t let the bad guys steal democracy
As a foreigner, it’s been a thrill to watch the events leading up to the US elections unfold. As a writer focused on AI, not so much.
Lest I become a fear monger, let me say that no one needs vanguard innovation to spread lies and issue convincing propaganda. Standard channels of distribution like social media matter more than ChatGPT and they’re not new.
I will admit, however, that text and image AI tools that came after 2022 can turn an election day — and months of pre-election campaign — into the worst nightmare for democracy advocates. Algorithms, old and novel, paint misty the prospect that awaits us in the months before the US election on November 5th.
Or any election, anywhere, for that matter.
We didn’t prevent this risk because one, tech companies were too busy making money, and two, in the last few years AI evolved way faster than we were prepared for. In some ways, it’s still quite dumb (e.g. trying to win at Tic-tac-toe), but it’s good enough at telling stories and drawing pictures to upend an otherwise fair process.
This election is the first time we face the danger of post-ChatGPT AI in a high-stakes sociopolitical scenario.
