Deep fake

As explained in Introduction to Generative AI, Second Edition

A deepfake is synthetic or manipulated media generated with machine-learning techniques to depict people, events, or statements inaccurately or deceptively.

A deepfake is synthetic or manipulated media created with artificial intelligence to make people appear to say or do things inaccurately or deceptively. The term combines “deep learning” and “fake” and originally described face-swapped videos, especially fabricated depictions of particular people. Its meaning has broadened to include photorealistic fictional people, cloned voices, fabricated recordings, and propaganda-like content. Deepfakes are one part of the wider category of synthetic media, which also includes AI-generated text, images, video, voice, and data. section 4.1

Deepfakes create both creative possibilities and serious social risks:

• Expand access to professional-looking images, video, and audio for creators without large budgets or specialized teams. section 4.3
• Enable impersonation, surveillance, identity theft, privacy violations, political manipulation, and disinformation. section 4.3 section 10.12
• Produce a liar’s dividend: people can dismiss authentic evidence by falsely labeling it a deepfake. section 4.3
• Support protective uses, such as anonymizing activists’ faces while preserving the emotional force of documentary footage. section 4.3

From fabricated announcements to cloned voices

Examples include a fabricated video of President Biden announcing a military draft, a manipulated clip of Mark Zuckerberg, AI-generated explicit images of Taylor Swift, election disinformation using synthetic speech, and cloned relatives’ voices used in WhatsApp scams targeting older people. section 4.1 section 10.12

Figure 4.1 The landscape of synthetic media

Maps deepfakes within the broader landscape of synthetic media, including generated or modified text, images, video, voice, and data. section 4.1

How Introduction to Generative AI, Second Edition covers it

How deep it goes

Defines the category and explains social effects, but stops short of technical generation methods or detection procedures.

Examples

Uses fabricated political videos, manipulated celebrity media, election robocalls, cloned-voice scams, and documentary face anonymization.

What you can do after

Evaluate deepfakes as creative tools and sources of deception, privacy harm, weakened evidence, and institutional distrust.

How it connects

Connects deepfakes to synthetic media, generative AI, privacy, identity theft, misinformation, journalism, legal systems, and public trust.

liveBookmanning.com

Read Introduction to Generative AI, Second Edition on liveBook Get the book at manning.com

Also explained in

• 

  liveBookmanning.com
  Sutskever's ListA deepfake is realistic AI-generated synthetic media depicting people doing or saying things that did not occur.
• 

  liveBookmanning.com
  Build a Text-to-Image Generator (from Scratch)A deepfake is synthetic or manipulated media generated or altered with machine-learning methods to impersonate or misrepresent real people or objects.
• 

  liveBookmanning.com
  Inside AIA deepfake is digitally generated or altered media that realistically depicts a person or event in a way that is not authentic.

Compare how 4 books explain deep fake →

Related concepts

• deep learninguses · Deep learning is a machine-learning approach that uses neural networks with multiple computational layers to learn representations from data.
• artificial intelligenceproduced by · Artificial intelligence is the field concerned with creating computational systems that perform tasks associated with human intelligence, such as learning, reasoning, perception, or decision-making.
• fabricated contentproduces · Fabricated content is media or information that has been artificially created or altered to appear authentic despite not representing genuine events or facts.
• generative modelingproduced by · Generative modeling is a machine-learning approach that learns patterns or a data distribution to generate new samples resembling the data.
• deepfake detectionaddressed by · Deepfake detection is the analysis of digital media to determine whether it has been synthetically generated or manipulated.
• generative adversarial networkproduced by · A generative adversarial network is a generative model with a generator and discriminator trained in competition, enabling the generator to produce samples resembling a target dataset or distribution.

This explainer was generated by Manning's knowledge graph from the text of Introduction to Generative AI, Second Edition. The section links point to the parts of the book it draws on.