Deepfakes in Court: How Would You Prove a Video Isn't You?

If someone made a fake AI video of you, how would you prove it wasn't you? This week, the federal committee that writes the rules for evidence in U.S. courts heard from experts, including OpenAI and Meta, about deepfakes. Host James Henderson explains what happened, what is still only a proposal, and how real evidence can be protected.

We also cover five more AI stories from Oct 15–16, 2026: New York City's Grok hearing, China's proposed AI export controls, G7 cyber drills, new European guidelines for AI in science, and what the IMF says AI means for everyday workers and small businesses.

CHAPTERS

0:00 Real or AI fake?

1:37 Story 1: U.S. courts face the deepfake problem

5:55 Story 2: New York City questions Grok's safety

9:17 Story 3: China proposes new limits on AI technology

12:03 Story 4: Major economies prepare for AI cyberattacks

14:53 Story 5: Europe updates guidance for AI in science

17:25 Story 6: AI and the global economy (IMF)

19:33 The takeaway: it's about trust

21:21 Your turn

KEY FACTS (checked against the sources below)

- The Advisory Committee on Evidence Rules met Oct 15, 2026 in Boston and held an expert panel on deepfakes.

- The judge survey (15 of 931 responding judges had faced a deepfake challenge) is a Federal Judicial Center survey released in March 2026.

- Proposed Rule 707 was withdrawn for revision in May 2026. A draft Rule 901(c) on deepfakes has not been approved. Neither is a federal rule.

- On-screen corrections: the NYC Council Speaker is Julie Menin. The company is SpaceXAI (formerly xAI). The IMF Managing Director is Kristalina Georgieva. China's step already in effect is a pre-export registration requirement; the 40-technology list is still a proposal (comments until Nov 15, 2026).

SOURCES

U.S. Courts, Advisory Committee on Evidence Rules agenda book, Oct 15, 2026: https://www.uscourts.gov/sites/default/files/document/2026-10_evidence_rules_agenda_book_9-23_final.pdf

Reuters via IBTimes UK, OpenAI warns courts on deepfake evidence, Oct 16, 2026: https://ibtimes.co.uk/openai-ai-deepfake-evidence-courts-1826007

Esquire Deposition Solutions, Federal judges and deepfake evidence challenges (FJC survey): https://www.esquiresolutions.com/federal-judges-set-bar-for-deepfake-evidence-challenges/

New York City Council, SpaceXAI agrees to testify following subpoena, Oct 13, 2026: https://council.nyc.gov/press/?p=3295

La Voce di New York, Menin and Grok hearing, Oct 16, 2026: https://lavocedinewyork.com/en/new-york/2026/10/16/spacexsi-cant-explain-how-menin-got-risque-grok-image-of-herself-in-a-second/

Reuters via KFGO, Beijing seeks tighter grip on AI exports, Oct 16, 2026: https://kfgo.com/2026/10/16/china-proposes-adding-40-ai-biotech-solar-technologies-to-export-control-list/

Reuters via KFGO, G7 to launch exercises to guard against AI cyber attacks, Oct 15, 2026: https://kfgo.com/2026/10/15/g7-to-launch-exercises-to-guard-against-ai-cyber-attacks/

European Commission, ERA Living Guidelines on generative AI in research (updated Oct 16, 2026): https://research-and-innovation.ec.europa.eu/news/all-research-and-innovation-news/era-living-guidelines-responsible-use-generative-ai-research-updated-2026-10-16_en

Reuters via Daily Sabah, IMF chief on AI safeguards, Oct 16, 2026: https://www.dailysabah.com/business/economy/ai-risks-may-exceed-opportunities-without-safeguards-imf-chief

DISCLOSURE

This episode was edited with AI assistance (Claude by Anthropic). Graphics and captions were checked against the sources above. Corrections are shown on screen. James's words were not changed. Coffee is not a paid sponsor.

#Deepfakes #AIethics #ArtificialIntelligence

More from the AI Coalition Network: https://ai-coalition.net

Leave a comment: how would you prove a video of you was fake?