• Pascal's Chatbot Q&As
  • Posts
  • Plaintiffs claim that Photobucket sold their images to third parties for purposes such as creating biometric facial recognition databases and training generative AI, without proper consent​​.

Plaintiffs claim that Photobucket sold their images to third parties for purposes such as creating biometric facial recognition databases and training generative AI, without proper consent​​.

Photobucket’s tactics to amend terms of service—such as coercive emails to reactivate accounts and enforce agreements—were described as misleading and a breach of trust.

Question 1 of 3 for ChatGPT-4o: Please analyze the article "Photobucket users and subjects sue to stop images from being sold as deep fake fodder" and the associated court document and list all the grievances put forward by the plaintiffs

ChatGPT-4o’s answer:
From the analysis of the article and court document, the grievances brought forward by the plaintiffs against Photobucket and unnamed defendants include:

  1. Unauthorized Sale of Images:

    • Plaintiffs claim that Photobucket sold their images to third parties for purposes such as creating biometric facial recognition databases and training generative AI, without proper consent​​.

  2. Violation of Privacy:

    • Photobucket allegedly captured and used biometric information (e.g., facial geometry) from photographs without informed consent, violating privacy laws in multiple states like Illinois (under BIPA), California, Virginia, and New York​​.

  3. Deepfake Risks:

    • The plaintiffs expressed concerns that their likenesses could be used to create deepfakes or other AI-generated content that might misrepresent them.

  4. Deceptive Practices:

    • Photobucket’s tactics to amend terms of service—such as coercive emails to reactivate accounts and enforce agreements—were described as misleading and a breach of trust. Users were misled into consenting to terms they might not fully understand​​.

  5. Breach of Contract:

    • Plaintiffs argue that Photobucket violated its initial user agreements by retroactively imposing terms that users did not consent to when they originally uploaded their photos​​.

  6. Exploitation of Dormant Accounts:

    • Photobucket’s use of dormant accounts to license content to third parties is contested. Many users had not engaged with the platform for years, and their silence or inactivity was deemed as implied consent by Photobucket​​.

  7. Failure to Address State-Specific Privacy Laws:

    • Despite knowing that some users resided in states with stricter biometric data laws, Photobucket allegedly failed to ensure compliance with these state-specific laws before using or licensing data​.

  8. Non-Consensual Use of Non-Account Holders' Data:

    • Plaintiffs also include individuals depicted in photos uploaded by others who never had accounts with Photobucket. These individuals were not bound by any agreement but were similarly affected​.

  9. Lack of Informed Consent Mechanisms:

    • The updated terms and policies did not adequately inform users about the extent of data use, including AI training purposes, nor did they allow meaningful opt-out options for affected users​.

  10. Emotional and Reputational Harm:

    • Plaintiffs are concerned about potential misuse of their likenesses, leading to emotional distress and possible reputational harm​.

  11. Irretrievability of Data:

    • Photobucket admitted it could not retrieve data once sold or shared with third parties, compounding the harm experienced by plaintiffs​.

These grievances form the basis of the plaintiffs’ request for injunctive relief, damages, and statutory penalties for the alleged violations. Let me know if you need a deeper analysis or summary.


Question 2 of 3 for ChatGPT-4o: What's the evidence presented and is it any good, do you think?

ChatGPT-4o’s answer:
Evidence Presented

  1. Emails from Photobucket:

    • Plaintiffs provided copies of emails sent by Photobucket to account holders. These emails prompted users to log in, "reactivate" accounts, or make decisions about their content. Clicking the links reportedly led users to accept updated terms of service, including a biometric policy​.

  2. Terms of Service and Policy Changes:

    • Photobucket’s earlier terms of service emphasized user rights over intellectual property and prohibited the use of content for unauthorized purposes. Plaintiffs argue that the updated terms retroactively granted Photobucket rights to use uploaded photos for AI and biometrics, violating the original agreements​.

  3. Statements from Photobucket:

    • Public statements from Photobucket, including its rebranding as an AI-focused company and its description of the "opportunity" presented by its massive photo database, are cited as evidence of its intent to monetize user content​.

  4. Plaintiff Testimonies:

    • Plaintiffs detailed their use of Photobucket, their lack of interaction with the platform in recent years, and their concerns about how their photos are being used without their consent​.

  5. Violations of Privacy Laws:

    • Evidence highlights potential violations of state laws, such as Illinois’ Biometric Information Privacy Act (BIPA), with specific provisions prohibiting unauthorized use of biometric data​.

  6. Lack of Informed Consent:

    • The plaintiffs emphasized that the methods used by Photobucket to obtain consent (e.g., email campaigns and "negative option" consent tactics) were coercive and insufficient under state laws​.

  7. Non-Compliant Licensing Practices:

    • Plaintiffs allege that Photobucket has already licensed their images and biometric data to unknown third parties without any ability to trace or retract the data​.

  8. Failure to Notify or Comply with Local Laws:

    • The lawsuit details Photobucket’s failure to comply with state-specific biometric and privacy laws, despite being aware of such requirements​.

Strengths of the Evidence

  1. Clear Violations of Privacy Laws:

    • The alleged violations of specific privacy statutes like BIPA are supported by detailed references to how these laws were violated (e.g., lack of informed consent). These statutes provide a strong legal framework for the plaintiffs' claims.

  2. Consistency in Allegations:

    • Plaintiffs’ testimonies are aligned, reflecting similar patterns of behavior by Photobucket. This uniformity strengthens their credibility and demonstrates systemic issues.

  3. Documented Policy Changes:

    • The plaintiffs have highlighted specific changes to Photobucket’s terms of service that seem to retroactively impose new obligations on users. This is a strong basis for claims of breach of contract.

  4. Photobucket’s Admissions:

    • Photobucket’s statements about selling or licensing biometric data and its inability to retrieve such data add credibility to the plaintiffs’ claims.

Weaknesses of the Evidence

  1. Lack of Direct Evidence of Harm:

    • The plaintiffs focus on potential harm (e.g., misuse of biometric data for deepfakes), but there is limited evidence of actual harm caused to individuals.

  2. Ambiguity in Consent Processes:

    • Photobucket might argue that users were provided opportunities to opt-out or that the terms were visible and accessible, though plaintiffs dispute the adequacy of these processes.

  3. Difficulty in Proving Licensing:

    • While the plaintiffs assert that data has been sold to third parties, they rely on Photobucket’s statements and do not provide specific evidence of these transactions or the third-party uses of the data.

  4. Challenges with Dormant Users:

    • Many plaintiffs had not interacted with Photobucket in years, which could complicate arguments about their reliance on older terms of service.

Overall Assessment

The evidence is solid in establishing a pattern of questionable practices by Photobucket, particularly in its handling of biometric data and terms of service updates. The strongest elements relate to violations of specific privacy statutes like BIPA, where strict standards for informed consent exist.

However, the case would benefit from:

  • Specific Examples of Harm: Demonstrating actual misuse of images or biometric data could strengthen the plaintiffs’ claims.

  • Third-Party Licensing Evidence: More concrete proof of who received the data and how it was used would provide a stronger foundation for damages.

The evidence appears credible and sufficiently detailed to establish a prima facie case, but the outcome may hinge on the court’s interpretation of informed consent and the enforceability of Photobucket’s revised terms of service.



Question 3 of 3 for ChatGPT-4o: What actions should Photobucket have taken and what should it be doing now?

ChatGPT-4o’s answer:
Photobucket's alleged practices highlight significant legal and ethical lapses, especially in handling user content and personal data. Here’s what the company should have done and what actions it should take now to mitigate harm and regain trust.

Actions Photobucket Should Have Taken

  1. Transparent Consent Processes:

    • Clearly disclose how user content might be used for AI, biometric, or other purposes.

    • Obtain explicit, informed, and written consent before using photos for purposes beyond original agreements.

    • Avoid “negative option” tactics (implied consent from non-response), which are widely criticized and may violate privacy laws.

  2. Comply with Privacy Laws:

    • Follow state-specific laws such as Illinois’ Biometric Information Privacy Act (BIPA), California Consumer Privacy Act (CCPA), and others.

    • Provide mechanisms to confirm and verify user residency to avoid violating local privacy statutes.

  3. Opt-In Instead of Opt-Out:

    • Implement an opt-in system for any new use of user data. Users should explicitly agree before their data is included in licensing or AI training initiatives.

  4. Respect Original User Agreements:

    • Avoid retroactive amendments to terms of service without explicit user approval.

    • Uphold original promises about intellectual property and data rights.

  5. Notification and User Control:

    • Notify users directly and clearly about any changes in data usage policies.

    • Allow users to control their data (e.g., delete, opt-out) without coercion.

  6. Limit Scope of Licensing:

    • Ensure any licensing agreements exclude sensitive uses like biometric processing or deepfake creation unless explicitly authorized by users.

  7. Data Deletion Protocols:

    • Develop transparent procedures to delete user content permanently upon request.

  8. State-Specific Compliance:

    • Adapt terms and processes to comply with the most stringent applicable laws, ensuring protection for all users.

Actions Photobucket Should Take Now

  • Stop all data sales and licensing immediately until explicit, informed consent is obtained.

  • Conduct a legal audit of practices to ensure compliance with privacy and biometric laws.

  • Work with regulators in jurisdictions like Illinois (BIPA) to resolve violations and avoid further penalties.

2. Transparent Communication

  • Issue clear statements to users explaining:

    • The extent of past data usage.

    • Actions being taken to safeguard their rights.

  • Apologize publicly for lapses and commit to improvements.

3. Implement a Remediation Plan

  • Revoke improperly licensed data:

    • Attempt to recall user data already shared with third parties.

    • If irretrievable, disclose the details (which companies received the data and for what purposes).

  • Provide opt-out mechanisms:

    • Allow users to opt-out retroactively and ensure their data is no longer used or licensed.

  • Update policies to explicitly require opt-in consent for any data usage.

  • Notify dormant users separately, allowing them to affirmatively agree to new terms or request data deletion.

5. Data Governance and Safeguards

  • Introduce a robust data governance framework:

    • Encrypt sensitive data to prevent misuse.

    • Limit internal and external access to user data.

  • Appoint a Data Protection Officer (DPO) to oversee compliance and liaise with regulators.

6. Engage Independent Oversight

  • Hire independent auditors to assess and certify compliance with data protection laws.

  • Share results with users and stakeholders for transparency.

7. Offer User Compensation

  • Provide affected users with compensation for unauthorized data usage or licensing.

  • Cover costs for identity theft protection services if biometric data was compromised.

8. Improve Platform Trust

  • Introduce a user-friendly dashboard where users can:

    • View how their data is being used.

    • Manage privacy settings.

    • Request deletion or withdrawal from data use.

9. Train Employees on Privacy Laws

  • Educate staff on data privacy, security, and ethical handling of user content to prevent future issues.

10. Collaborate with Industry and Advocacy Groups

  • Partner with privacy advocates and industry leaders to align practices with best standards.

  • Contribute to initiatives addressing responsible AI and content usage.

Conclusion

By taking these steps, Photobucket can begin to address the grievances raised and build trust with its users. The company must demonstrate a strong commitment to legal compliance, user rights, and ethical data practices to avoid further legal challenges and reputational damage.