Should we consider a class action lawsuit against Shutterstock?

Home Forums Commentary Should we consider a class action lawsuit against Shutterstock?

Viewing 9 posts - 1 through 9 (of 9 total)
  • Author
    Posts
  • #46517
    Music1234
    Participant

    A great majority of us have contributed music to Pond5/ Shutterstock over the last 15 years.

    In the beginning it was great. We licensed a track and split the royalty 50/50 with Pond5.
    Then three years ago they shifted the non exclusive payments to 65%/35%.
    Then they introduced subscription royalties.

    But April 2023 was a major change that caught us all off guard. We sudenly saw “dataset revenue” on the statement.

    What exactly is a dataset? In a nutshell it’s when one company offers an entire massive dataset (i.e. all of our music catalogs, wave files, titles, descriptions of tracks, keywrods of tracks, etc.) to another company for copying and ingestion to train AI learning models.

    To my knowledge, none of us ever were issued an “opt in” notification by Shutterstock during 2022 where we all willingly agreed to allow our property to be copied and ingested by AI learning models that Meta and Open AI have developed. Instead, Shutterstock simply made the deals with Meta and Open AI, then deposited some tip money in our accounts and made that bold statement “you can always opt out if you want to?”

    Was this legal?

    Was there a breach of copyrights in these bold actions licensing all of our data to META and OPEN AI?

    I think yes.

    Not only did they act unilaterally without permission from copyright holders, they then promised that “dataset revenue” would post quarterly moving forward. The first line item for “Dataset Revenue” occurred on April 12, 2023. Since then, to my knowledge no one has received a penny more, yet when you read articles about Shutterstock’s finacial press releases, they continue to tell their investors/ stockholders that dataset revenue is growing and making up a huge portion of their 2023 revenue and they expect even more growth in 2024. The question remains though: Was this even a legal action that they took licensing others music property for AI training, without obtaining legal permission from rightsholders by way of “opt In” notification to the new terms of service for sellers?

    I look forward to everyone’s thoughts.

    #46519
    Art Munson
    Keymaster

    Bump

    #46520
    Music1234
    Participant

    Here are links to some key Press releases. I suppose we can add LG corporation to the list they are making lucrative AI Training deals with. The problem is that none of this revenue is being shared with rights holders in 2024. Yet, contractually speaking, they are obliged to pay contributors for AI training. “Support employees” for Shutterstock do not have a valid answer as to when the next round of AI Training earnings may post to our accounts.

    https://investor.shutterstock.com/news-releases/news-release-details/shutterstock-partners-openai-and-leads-way-bring-ai-generated

    https://investor.shutterstock.com/news-releases/news-release-details/shutterstock-joins-forces-lg-ai-research-advance-ai-technology

    Shutterstock symbol SSTK is going to post earnings and host a conference call on August 6th, 2024

    https://investor.shutterstock.com/news-releases/news-release-details/shutterstock-report-second-quarter-2024-earnings-results-august

    I suggest that everyone call in and listen, maybe even ask some questions on the conference call, Shutterstock essentially has a monopoly on stock media at this point. They control a lot of this market. So much that it is dangerous how much intellectual property they represent for licensing.

    #46550
    Advice
    Participant

    I’m not a lawyer but AFAIK, class action lawsuits are expensive and complex, require a very large number of participants, with the litigants usually ending up with a trivial amount of money each. I’ve been the recipient of settlements of $3, 3 months free subscription to something, etc. The legal fees eat up the money.

    It would seem to be that if you are dissatisfied with P5/Shutterstock (and you have reason to be so), the best thing to do is to simply pull all your tracks and move on.

    JMHO. That doesn’t at all mean I condone what P5/Shutterstock is doing and I will most likely shut my account down.

    #46552
    Music1234
    Participant

    Legally speaking Shutterstock/ P5 did not comply with laws which explicitly state that sellers/ Licensors (Music Producers in this instance)) of intellectual property must first be asked to OPT IN with very clear understanding of what the goals are and were with licensing datasets to the largest tech companies like META, OPEN AI, GOOGLE, MICROSOFT, AMAZON, LG, etc…

    You do not just send contributors of intellectual property a payment and then say “oh by the way you can always opt out of dataset deals.” There has been a major breach of compliance to the terms of service at that moment in time back in late 2022 early 2023. I don’t think any of us knew what “datasets” were, nor were we properly informed prior to the dataset transaction being made, nor were we asked to opt in with clear understanding of how these deals would be structured.

    Frankly, we still don’t know anything at all about what they were paid first, and how they decided to share the revenue. I think any music producer who was dealing on the platform at that time has a very strong case against Shutterstock. Prior case law and compliance norms in respect to “opting in” would favor the argument that one must first willingly agree to OPT IN as opposed to being “unilaterally opted in”, then paid, then told one can OPT OUT after the fact.

    This was an extremely risky move that Shutterstock/ p5 made 16 to 20 months back and frankly, we have no idea how they are dealing the music (or datasets) in 2024 nor how much? when? and if music producers will be fairly compensated

    #46568
    Music1234
    Participant

    Listen to Shutterstock’s conference call carefully

    https://edge.media-server.com/mmc/p/fffgc3rf/

    Key Takeaways

    121% revenue growth licensing data in Q2 – It’s profitable to license everyone’s IP but not pay the IP owners…right?

    They have 3 types of deals they are engaging in to license our “data” (music files, titles, and keywords)

    1. Term license deals
    2. Large Corpus deals where they enter in partnerships where they receive cash and equity in new AI companies.
    No revenue is recived but a revenue sharing product is being created with NVIDA and Data Bricks
    3.They have engaged in Term License deals for 3 to 5 years so the customers get broad based access to the data for the express purpose of training AI models.

    They state: “Accounting of these deals can vary depending on the structure of the deal. Some get paid right away upon transfer of the IP (intellectual property) over the few initial quarters.

    Cue to 18:00 in and listen.

    They stated at 20:00:

    our contributor community is compensated in each and every structure in line with our ethical approach to licensing data for generative AI

    Ha! Laughable! Where is the compensation? There has been no compensation for 16 months!. I highly recommend that everyone write to customer support at all shutterstock properties to ask when and how they will be compensated for these data deals where our IP is being handed over to companies like NVidia, META, and other “small and large companies wanting access to the data”.

    Guys, this is just highway robbery here. None of this is being regulated at the moment by governments, but it’s clear as day from this conference call that they are giving away our IP to major players in the AI space like NVIDIA, MICROSOFT, META, etc….and profiting a lot from it! Our data is training generative AI music models

    What exactly is “in line with our ethical approach to licensing data for generative AI”?
    What exactly is the “ethical approach”?

    As far as I can see, they are just licensing data and booking the profits for themselves.

    They even stated that these deals can bring in 25 to 50 million dollars! Now I know not all of this is for music files, but certainly some of it is. Please write to support and ask where your piece of the pie is today. I’m the only one talking here. Are you all just fine and OK with the way this is playing out? Your music is getting sold/ licensed to NVIDIA and META and MICROSOFT and OPEN AI (To train their AI MODELS), but you are not getting paid.

    #46629
    ENW1
    Participant

    I was in Pond 5 for several years. Then, they began to squeeze us to take a bigger piece of the pie. I recall them announcing they planned to offer our Tracks to train AI. They did NOT offer an Opt In/Out. That’s when I dumped them. I’ve tightened-up all of my agreements. Fractions of Pennies won’t due. Now, I won’t deal with anyone unless it’s worth my time.

    #46912
    MaxPower
    Participant

    I think you may have frightened them Music1234 – I’ve had over fourteen hundred dollars appear in my account as ‘dataset earnings’!
    Thank you !

    #46918
    TechNoiZ
    Participant

    I noticed these ‘earnings’ earlier this year but as much as I tried I could not update an email address on my own account to claim them. They site there still … But I agree, absolute daylight robbery with no permission sought … what a surprise!!!!

Viewing 9 posts - 1 through 9 (of 9 total)
  • You must be logged in to reply to this topic.
X

Forgot Password?

Join Us