
<lom:lom xmlns:lom="https://oer-repo.uibk.ac.at/lom" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="https://oer-repo.uibk.ac.at/lom/latest https://w3id.org/oerbase/profiles/lomuibk/latest/schemas/lom-uibk.xsd">
  
<lom:general>
  
<lom:identifier>
  
<lom:catalog>door.donau-uni.ac.at</lom:catalog>

  
<lom:entry>
  
<lom:langstring xml:lang="x-none">o:5615</lom:langstring>

  
</lom:entry>

  
</lom:identifier>

  
<lom:identifier>
  
<lom:catalog>DOI</lom:catalog>

  
<lom:entry>
  
<lom:langstring xml:lang="x-none">10.48341/dx7k-e072</lom:langstring>

  
</lom:entry>

  
</lom:identifier>

  
<lom:title>
  
<lom:langstring xml:lang="en">Restoring Trust in the Age of Deepfakes: A Blockchain-Based Proposal : The Global Challenge of AI-Manipulated Media</lom:langstring>

  
</lom:title>

  
<lom:description>
  
<lom:langstring xml:lang="en">The proliferation of AI-generated deepfake media represents a critical global challenge, undermining 
public trust across political, financial, and social domains. Recent advances in generative AI, particularly 
Google’s VEO3 model, have dramatically escalated the sophistication and accessibility of synthetic media 
creation, enabling the generation of photorealistic videos with synchronised audio that are nearly 
indistinguishable from authentic content. Existing approaches, primarily focused on deepfake detection, 
fall short due to the evolving sophistication of artificial intelligence methods and the emergence of what 
researchers term the “liar’s dividend”—the ability for bad actors to dismiss authentic content as 
potentially fake. This opinion paper introduces a technically feasible solution: the optional integration of 
cryptographic hashing, secure metadata anchoring (including GPS coordinates and device specifications), 
and optional digital identity signatures (eIDAS, EUDIS, or decentralised Self-Sovereign Identity 
frameworks) directly into media-capturing devices such as smartphones and cameras. These authenticity 
markers would be immutably anchored to public blockchain infrastructures—examples include Ethereum 
Layer 2 solutions, Solana, and Ardor—creating an incorruptible provenance ledger. The article examines 
the multiple advantages of such blockchain-based media authentication, including strengthened content 
integrity, clear provenance, rapid verification capabilities, deterrence of misinformation, protection of 
intellectual property, and resilience across decentralised platforms. Despite its promise, the 
implementation faces challenges related to privacy and anonymity concerns, device integration 
complexities, scalability, verification infrastructure, and evolving legal and regulatory landscapes. To 
overcome these barriers, parallel initiatives in user and stakeholder education are crucial, emphasising 
media literacy, transparency, and global inclusivity. By collaboratively addressing these issues, 
blockchain-based media authentication has the potential to significantly mitigate the impact of deepfakes 
and rebuild digital trust in society.</lom:langstring>

  
</lom:description>

  
<lom:language>eng</lom:language>

  
<lom:keyword>
  
<lom:langstring xml:lang="en">Blockchain Media Authentication</lom:langstring>

  
</lom:keyword>

  
<lom:keyword>
  
<lom:langstring xml:lang="en">Deepfake Detection</lom:langstring>

  
</lom:keyword>

  
<lom:keyword>
  
<lom:langstring xml:lang="en">Digital Provenance</lom:langstring>

  
</lom:keyword>

  
<lom:keyword>
  
<lom:langstring xml:lang="en">Synthetic Media</lom:langstring>

  
</lom:keyword>

  
<lom:keyword>
  
<lom:langstring xml:lang="en">Content Integrity Verification</lom:langstring>

  
</lom:keyword>

  
<lom:keyword>
  
<lom:langstring xml:lang="en">Cryptographic Timestamping</lom:langstring>

  
</lom:keyword>

  
<lom:keyword>
  
<lom:langstring xml:lang="en">Media Forensics</lom:langstring>

  
</lom:keyword>

  
<lom:keyword>
  
<lom:langstring xml:lang="en">Information Authenticity</lom:langstring>

  
</lom:keyword>

  
</lom:general>

  
<lom:lifecycle>
  
<lom:datetime>2025-11-25T08:10:13.377Z</lom:datetime>

  
<lom:contribute>
  
<lom:role>
  
<lom:source>
  
<lom:langstring xml:lang="x-none">LOMv1.0</lom:langstring>

  
</lom:source>

  
<lom:value>
  
<lom:langstring xml:lang="x-none">Author</lom:langstring>

  
</lom:value>

  
</lom:role>

  
<lom:centity>
  
<lom:vcard>BEGIN:VCARD
VERSION:3.0
N:Pfeiffer;Alexander;
FN:Alexander Pfeiffer
X-ORCID:https://orcid.org/0000-0002-8689-3318
END:VCARD</lom:vcard>

  
</lom:centity>

  
<lom:centity>
  
<lom:vcard>BEGIN:VCARD
VERSION:3.0
N:Krishna;Nanditha;
FN:Nanditha Krishna
X-ORCID:https://orcid.org/0000-0001-5536-4993
END:VCARD</lom:vcard>

  
</lom:centity>

  
</lom:contribute>

  
</lom:lifecycle>

  
<lom:educational>
  
<lom:learningresourcetype>
  
<lom:source>
  
<lom:langstring xml:lang="x-none">https://w3id.org/kim/hcrt/scheme</lom:langstring>

  
</lom:source>

  
<lom:id>https://w3id.org/kim/hcrt/text</lom:id>

  
<lom:entry>
  
<lom:langstring xml:lang="de">Textdokument</lom:langstring>

  
<lom:langstring xml:lang="en">Text</lom:langstring>

  
</lom:entry>

  
</lom:learningresourcetype>

  
</lom:educational>

  
<lom:rights>
  
<lom:copyrightandotherrestrictions>
  
<lom:source>
  
<lom:langstring xml:lang="x-none">LOMv1.0</lom:langstring>

  
</lom:source>

  
<lom:value>
  
<lom:langstring xml:lang="x-none">yes</lom:langstring>

  
</lom:value>

  
</lom:copyrightandotherrestrictions>

  
<lom:description>
  
<lom:langstring>https://rightsstatements.org/vocab/InC/1.0</lom:langstring>

  
</lom:description>

  
</lom:rights>

  
<lom:technical>
  
<lom:format>application/pdf</lom:format>

  
<lom:size>551406</lom:size>

  
<lom:location>https://door.donau-uni.ac.at/o:5615</lom:location>

  
<lom:thumbnail>
  
<lom:url>https://door.donau-uni.ac.at/api/object/o:5615/thumbnail</lom:url>

  
</lom:thumbnail>

  
</lom:technical>

  
<lom:classification>
  
<lom:purpose>
  
<lom:source>
  
<lom:langstring xml:lang="x-none">LOMv1.0</lom:langstring>

  
</lom:source>

  
<lom:value>
  
<lom:langstring xml:lang="x-none">discipline</lom:langstring>

  
</lom:value>

  
</lom:purpose>

  
</lom:classification>

  
</lom:lom>


