Ahead of the U.S. election, some analysts worried that artificial intelligence could imperil election integrity. Although AI did not end up disrupting the vote, experts aren’t writing off the risks it poses to democracy.
“I think it would be foolhardy to say: ‘Well, there’s been no major disaster yet, so we’re okay here,’” Gary Marcus, a scientist and AI expert, recently told FP’s Rishi Iyengar. “That’d be like saying we made a bunch of steamships, so this one’s invincible, and whoops, you hit an iceberg.”
In this edition of Flash Points, FP contributors consider the ways AI could endanger democratic societies and how policymakers might face down those threats.
U.S. President Joe Biden hands a pen to Vice President Kamala Harris after signing an executive order on artificial intelligence at the White House in Washington on Oct. 30, 2023. SIPA via Reuters
AI’s Alarming Trend Toward Illiberalism
Left ungoverned, the technology opens pathways to undermine democracy, Ami Fields-Meyer and Janet Haven write.
Álvaro Bernis illustration for Foreign Policy
What AI Will Do to Elections
Depleted tech platforms, AI-enabled misinformation, and countries going to the polls, FP’s Rishi Iyengar reports. What could go wrong?
People attend Computex 2024 in Taipei on June 4.I-Hwa Cheng/AFP/Getty Images
The Science of AI Is Too Important to Be Left to the Scientists
Concerted international action will require political will, Hadrien Pouget writes.
Mourners carry crosses bearing the names of victims of the Mugunga Camp bombing attack as they attend a funeral ceremony in Goma, Democratic Republic of the Congo, on May 15. Guerchom Ndebo / AFP
How Africa’s War on Disinformation Can Save Democracies Everywhere
African leaders can’t afford to wait for Big Tech. By taking action, the continent could spare future generations from the scourge of adversarial AI, Abdullahi Alim writes.
A woman views historical documents and photographs displayed in a high-tech art installation created with artificial intelligence, seen in Istanbul on May 6, 2017.Chris McGrath/Getty Images
Red Teaming Isn’t Enough
Researchers need far more information to understand AI’s true risks, Gabriel Nicholas writes.