Disinformation Security: Moon Hoax to AI Deepfakes | Ep. 2026.30

From Bits & Bytes & Business · 00:51:29

Show by Joseph H. Schuessler, Ph.D.

In August 1835, a New York newspaper told readers that an astronomer had seen bison, unicorns, and bat-winged people on the Moon. None of it was real, and the paper kept right on selling. This week's Bits & Bytes & Business follows that hoax, link by link, to the AI deepfake ads now running in the 2026 midterms.Along the way: the penny press and its advertiser-funded business model, the 1890s circulation war between Pulitzer and Hearst, broadcast ratings, and a 2018 MIT study of about 126,000 stories showing that falsehood was 70 percent more likely to be retweeted. Then Gartner's Disinformation Security category: its High benefit, 1% to 5% penetration, and Emerging ratings in plain words, and what it means for fraud, executive impersonation, and brand risk.The episode takes a clear position: lies pay when they cost almost nothing to make, when the system that spreads them is paid by attention, and when the penalty for getting caught is smaller than the payoff. We give the strongest counterarguments a real hearing, including satire as protected speech, and we describe tactics rather than naming candidates.00:00 Cold open: August 25, 183501:10 From penny paper to global feed01:34 Welcome and today's mission02:43 Recap of Episode 29: the inspector test04:14 The pivot: what if the inspected thing is information?04:49 The arsonist-paid fire inspector05:45 Back to 1835: the Moon Hoax revisited06:32 The reveal: Richard Adams Locke and the 1840 admission07:51 Benjamin Day's penny Sun: the reader becomes the product09:34 Did the hoax make the Sun's fortune? The legend10:31 The 1890s circulation wars and the Yellow Kid12:22 Broadcast ratings: same math, new medium13:07 The 2018 MIT study on false news14:44 It was not the bots15:46 The novelty hypothesis17:40 The system is working as designed18:16 The generative collapse: lying costs almost nothing19:15 How GANs and voice cloning work20:47 Gartner's Disinformation Security, in plain words21:47 Why adoption is only 1% to 5%22:51 Four layers of defense: provenance and C2PA24:15 Last week's White House AI accord24:46 Detection, identity, and narrative intelligence26:25 Devil's advocate: why the defenses lag28:17 Business voice compromise and fraud29:50 Deepfake red-teaming and the trust council31:14 The 2026 midterms: where lies cost nothing32:19 The Wesleyan Media Project numbers, as reported33:44 Tactics: fake conversations and the satire defense35:35 The disclosure paradox: 32% versus 29%36:32 The label design problem: NYU findings38:05 The position: why lies pay39:13 Four ways to change the payoff41:21 Pushback: free speech and the satire line43:24 Pushback: harmless uses and small campaigns44:42 Pushback: the arms race46:13 Closing the loop: the lens is free48:03 Recap and call to action48:43 The debate question49:21 Next week's puzzle: Bengal, 1858Companion resources:Blog post: https://www.linkedin.com/pulse/telescope-never-existed-joseph-h-schuessler-fddccVideo overview: https://youtu.be/a5EDt3fFBIEElection figures are as reported by the Wesleyan Media Project; see the blog post for sources. Anthropic, which makes Claude, is among the reported signatories of the White House AI accord mentioned in this episode.Like, follow, and subscribe so you do not miss next week's episode. Tell us in the comments: should a campaign be allowed to air an AI-made video of an opponent saying something they never said, even with a label?Bits & Bytes & Business is a weekly series produced by Dr. Joseph H. Schuessler of the Dr. Sam Pack College of Business at Tarleton State University. Developed with the assistance of AI tools, including Claude, Gemini, and NotebookLM, guided by human-authored prompts, editorial review, and external research.

More episodes of Bits & Bytes & Business

More episodes

PodSearch is a free, independent podcast search engine. Read about PodSearch, the privacy policy or contact us.