Deepfakes
Originally designed for use in pornography, deepfake technology has spilled over into other aspects of life. The above use of facial recognition in killer drones is just one negative use of the technology, but I would say it’s not the most pressing racial recognition problem we are facing. The next evolution of facial recognition technology are deepfakes. Which are videos or photots generated by an artificial intelligence. Deepfake videos were easy to spot years ago, they were usually grainy and the humanoids in the video clips didn’t look right. Deepfakes have come a long way since they first appeared some years ago. They’re good, but they are not yet indistinguishable from real people. That being said, it is important to look at the prospective dangers of this technology.
The current consensus on the dangers of deepfake technology mostly revolves around political elections and fake news. The fear is deepfake technology can be used to replicated political candidates and elected officials. Audio replication technology is the newest extension of deepfake technology, which made it progressively more complicated to differentiate actual audio and video recordings from deceptive misrepresentations created by manipulating real sounds and images. When this technology matures in the coming years, fake news and misinformation is going to explode, and nobody knows what to do about it.
Federal and state legislators should focus on combating the subsequent fake news and misinformation. The single most important thing governments can do is help educate the populace about this new technology. At the same time, governments should prioritize rebuilding the broke trust of the various institutions, bureaucrats, and elected officials. Congress should work with private technology companies to help fight deepfakes on their websites. Social media companies are the most at risk to deepfake technology, and government should assist them in any way they can to fight falsified audio and video recordings.
