The world’s largest tech companies have announced a partnership to tackle abuse of artificial intelligence in elections, they said on the sidelines of the Munich Security Conference on Friday.
The 20 companies – including OpenAI, Meta, Microsoft, Amazon, Google, TikTok and X – have agreed to share information on voter deception through AI. In particular, they pledge to “work on tools” to spot misleading content created by AI, so that users can identify it.
One idea is to affix a “digital tattoo” to videos that are generated by the AI tools developed by these companies, invisible to the naked eye but capable of being detected by a machine. “All solutions have limits,” warn the signatories, who also include Adobe, LinkedIn, Amazon and IBM.
"We have a responsibility to help ensure these tools don’t become weaponised in elections"
Generative AI makes it possible to create text, images, sound files or videos via a simple request in everyday language, which can make fake documents that appear authentic. Using archives, the technology can produce deepfakes, which show a person saying or doing something that did not happen.
“We have a responsibility to help ensure these tools don’t become weaponised in elections,” said Brad Smith, vice president of Microsoft.
The digital giants are under pressure to act against the spread of such content, with major elections scheduled for 2024 around the world, including in Belgium, the EU, US, UK, India and Russia.
“With so many major elections taking place this year, it’s vital we do what we can to prevent people being deceived by AI-generated content,” said Nick Clegg, president of global affairs at Meta.
Comments