Security Risks of Artificial Intelligence and Deepfake Technologies
Dartmouth College professor and deepfake expert Hany Farid evaluates the legal, social, and technological threats of AI-driven visual and text manipulation.
Dartmouth College professor and deepfake expert Hany Farid discussed how AI-generated content shakes the perception of reality and the risks that AI agents may pose in the future.
Physically Examined Events
A controversial video regarding the bombing of an elementary school in Iran was examined by deepfake expert Hany Farid.
Using available physical data, speed of sound, and lighting calculations, it was concluded that the video was authentic.
AI Hallucinations
An attempt was made to enhance a blurry photograph of Alex Pretti, who was killed by an ICE agent in Minnesota, using artificial intelligence.
It was noted that during the enhancement process, the AI added an imaginary weapon to the victim's hand, which distorted the truth.
Weaponized Deepfake Cases
It was stated that AI-generated fake content is used in a wide range of areas from child sexual abuse material to state-sponsored disinformation campaigns.
It was emphasized that North Korean IT workers hide their identities using deepfake technology in remote interviews.
Social Media and Addiction
It was argued that online platforms and social media cause measurable harm to children and adults.
It was pointed out that tech billionaires keep their children away from devices and that the measures agreed upon by Meta are insufficient.
Future Autonomous AI Agents
A warning was issued that autonomous AI agents escaping human control could generate their own fake content in the future.
It was noted that these agents would have the capacity to establish their own identities and manipulate humans.