LinkedIn’s New Generative AI Feature: What You Need to Know
Western Data, which creates the biases we see in AI. So, do we continue to
be excluded from AI models?
Regards
*Ali Hussein*
Fintech | Digital Transformation
Tel: +254 713 601113
Twitter: @AliHKassim
LinkedIn: Ali’s Profile <ke.linkedin.com/in/alihkassim>
<ke.linkedin.com/in/alihkassim>
Any information of a personal nature expressed in this email are purely
mine and do not necessarily reflect the official positions of the
organizations that I work with.
On Thu, Sep 19, 2024 at 2:48 PM Jacinta Wothaya via KICTANet <
[email protected]> wrote:
> Dear listers,
>
> LinkedIn has stirred up controversy by introducing a feature allowing the
> platform and its affiliates to use personal data and user-generated content
> to train generative AI models
> <www.linkedin.com/help/linkedin/answer/a6278444>. While this move
> reflects the growing trend of data commodification in the age of artificial
> intelligence, it raises serious concerns regarding user consent and
> privacy. The new feature allows LinkedIn to leverage the vast amount of
> data generated by its users to enhance its AI capabilities. This decision
> is not unexpected; as AI technology becomes more sophisticated, data is
> increasingly recognized as a valuable asset. However, LinkedIn’s
> implementation has come under fire for its lack of transparency. *Many
> users were automatically opted in to this feature without prior
> notification*, igniting fears over data misuse. The company has just
> updated the privacy policy on its website
> <www.linkedin.com/legal/privacy-policy#use> to reflect the new
> changes, effective September 18, 2024.
>
> According to LinkedIn’s FAQs
> <www.linkedin.com/help/linkedin/answer/a5538339>, opting out
> means that the platform and its affiliates won’t use your personal data or
> content to train models going forward. However, this does not affect any
> training that has already taken place. Furthermore, opting out does not
> prevent LinkedIn from using your personal data for training
> non-content-generating generative AI models. Users must object to this
> latter use by filling out a separate opt-out form provided by LinkedIn
> <nam06.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.linkedin.com%2Fhelp%2Flinkedin%2Fask%2FTS-DPRO&data=05%7C02%7Ctviano%40linkedin.com%7C1de2bb57c76840ca36da08dca1bff3c2%7C72f988bf86f141…>
> .
>
> The move appears to contravene several important regulations designed to
> protect user privacy. Under the General Data Protection Regulation (GDPR)
> in the EU, Article 6 stipulates that personal data must be processed
> lawfully, fairly, and transparently. LinkedIn’s failure to notify users may
> violate these principles, particularly the requirement for informed
> consent. Furthermore, Article 7 mandates that consent must be freely given
> and can be withdrawn at any time. LinkedIn’s FAQ for its AI training claims
> that it uses “privacy-enhancing technologies to redact or remove personal
> data” from its training sets. Notably, the platform states it does not
> train its models on users located in the EU, EEA, or Switzerland, which may
> provide some level of assurance for users in those regions.
> Similarly, the Kenya Data Protection Act (2019) emphasizes the importance
> of consent. Section 26 of this act requires data controllers to obtain
> explicit consent from users before processing their personal data. By
> automatically opting users in, LinkedIn could be infringing upon these
> legal protections, raising significant questions about its compliance with
> data protection laws.
>
> Notably, LinkedIn’s recent move isn’t an isolated case but is part of a
> broader trend where tech giants exploit user data to fuel AI advancements.
> Only recently, Meta allegedly confessed to using all public text and
> photos of adult Facebook and Instagram users to train its AI models since
> 2007
> <www.theverge.com/2024/9/12/24242789/meta-training-ai-models-facebook-instagram-photo-post-data>
> .
>
> Such practices raise important questions about user rights, data
> ownership, and ethical considerations in AI development. While the
> potential for innovation is significant, the risks associated with
> unauthorized data use cannot be overlooked. Tech giants will continue to
> push the boundaries of data utilization, and we are likely to see
> increasing scrutiny from governments and regulatory bodies worldwide.
> Nonetheless, existing laws may not be sufficient to address the
> complexities introduced by AI and big data, and the need for robust
> legislation to increase transparency, consent, and accountability in data
> usage has never been more pressing. At the moment, it is the user’s
> responsibility to stay informed and proactive about their data privacy but
> we look forward to a time when all tech companies innovate with user
> protection as the priority.
>
> *How to Opt-Out of Your Account Being Used for Training Generative AI *
>
> 1. While logged into your LinkedIn account, go to *Settings & Privacy*.
> 2. Click on *Data Privacy*.
> 3. Select *Data for Generative AI Improvement* and turn off the
> feature.
> 4. To stop your data from being used for non-content-generating AI
> models, complete the following form provided by LinkedIn
> <www.linkedin.com/help/linkedin/ask/TS-DPRO>.
>
>
> Best,
>
> *Jacinta Wothaya,*
> *Digital Resilience Fellow @**KICTANet* <www.kictanet.or.ke/>, @*tatua
> <tatua.digital/>*
> LinkedIn: *Jacinta Wothaya
> <www.linkedin.com/in/jacinta-wothaya-510a8b153>*
>
>
>
