Deepfake technology allows users to manipulate media using AI to produce fictional images, audio, and video. The term is a combination of "deep learning" and "fake," with the former referring to computers being taught to recognize complex patterns in data to produce realistic content.
Hours of research by our editors, distilled into minutes of clarity.
Unlike most content created from large language and diffusion models, deepfakes intentionally misrepresent real people or events with the intent to deceive. As the products of this technology become more realistic, they pose increasing risks to public trust, security, and privacy.
Artificial neural networks can identify an individual's physical characteristics when provided with training data, such as real footage of the person speaking. Graphics techniques overlay these with matching characteristics from another real or generated person to create novel footage of things that have never taken place in reality.
Advances in artificial intelligence have made deepfakes increasingly difficult to identify, as they are intentionally designed to appear real. The technology has exacerbated some of the problems brought on by digital editing tools like Photoshop.
A Freedom House report found that AI-powered disinformation campaigns and censorship tactics are spreading, with 16 countries utilizing the technology to shape online narratives or suppress dissent, as of 2023. In Venezuela, deepfake software was used to push pro-government propaganda with fake news anchors.
Synthetic media tools can map voices and faces to make individuals appear as someone else. Although they open the door to new opportunities in immersive education and entertainment, they also raise concerns about the misuse of personal identity.
Digital replicas of the deceased provide individuals with another way to process grief in a culture with longstanding traditions of communicating with ancestors. The emotional and ethical implications of "digital immortality" are complicated further by cases of actors impersonating deceased relatives during deepfake video calls to hide their deaths.
A cybersecurity company developed a tool that mimics a person's voice with no noticeable delay using only publicly available software, hardware and voice samples, allowing users to have seemingly authentic conversations while pretending to be someone else.
Though deepfakes are often synonymous with manipulation and bad intentions, some artists have found creative uses for the technology. In this example, Billy Joel depicts himself in various eras of his career in keeping with the song's nostalgic message.
As the Apollo 11 astronauts prepared for their journey to the moon, two speeches were drafted for President Richard Nixon: one to be delivered upon the astronauts' success and another to comfort the public in case of failure. This interactive explores the creation of a believable version of the latter.
Differences in the backgrounds of training data with the same subject introduce variability, resulting in background textures and text that lack realistic detail. Difficulties with managing long-distance dependencies, the complexity of hair, and other real-world phenomena introduce errors that can be used to flag AI-generated content.
Since the time our ancient human relatives began using stone tools to perform tasks, humans have harnessed scientific knowledge and new technologies to advance our ability to feed ourselves, increase our health, expand the boundaries of our knowledge, and much more. From artificial intelligence and quantum computing, to optimizing the human body in pursuit of a long and healthy life, and much more—learn about these topics and more with our smart-but-digestible overviews and expert-curated resources.