Tech Giants to Sign AI Accord Aimed at Labeling Deceptive Content, But Critics Question Impact
• Leading AI companies planning to sign "accord" to develop tech to identify and label deceptive AI content aimed at deceiving voters • Agreement does not ban political deepfakes, amounts to manifesto outlining risks and voluntary steps to mitigate them • Deepfakes can deceive public and jeopardize election integrity, companies agree • Tech companies already have policies on AI-generated political content, but broad system to identify & label content across platforms hasn't materialized • Adobe positioned itself as leader in reining in AI content, but its stock photo site recently filled with fake images of Gaza war