A lawsuit has been filed against Twitch and its parent company, Amazon, for using live broadcasts by content creators on Twitch’s streaming platform, to train AI models. This highlights a growing issue which is the unauthorised extraction of creator content to power artificial intelligence. The lawsuit follows reports that Amazon harvested live broadcasts, chat logs, clips, and audio to train generative AI models.
While tech platforms frame this data usage as a utility to develop higher-accuracy real-time subtitles and accessibility features, content creators argue it represents an uncompensated exploitation of their intellectual property. The controversy opens up a growing concern around consent, ownership, and the mechanics of platform governance.
Should streamers who dedicate thousands of hours cultivating unique online personas, voice signatures, and interactive communities be compensated for use of their content to train generative AI models? It will be interesting to follow this US lawsuit to see whether the outcome will create a precedent in how tech companies harvest data on their platforms to train AI models in the future.
The argument in favour of this lawsuit highlights that harvesting personal data, even under the guise of improving speech-to-text accuracy, turns intimate creator-fan interactions into raw material for commercial software to exploit. Creators argue that using live, unscripted content without prior explicit permission violates their expectation of control over their personal likeness and voice.
Rather than asking creators for permission upfront, platforms often activate AI training settings by default, shifting the burden onto the individual to opt out. This has been heavily criticised as the process for opting out is, in some cases, deliberately cumbersome to deter this process. In addition, many tech platforms will not set the default as opt out, believing that very few creators would choose to opt in in the current circumstances as they are effectively giving permission for free use of their content by these tech giants.
This lawsuit should act as a warning to the multitude of maritime tech platform and software providers. If they are ever tempted to use seafarer data, such as VoIP calls home, voice-activated bridge software, welfare streaming chat logs, or video calls, to fine-tune operational tools or train crew-welfare AI chatbots under the guise of “improving service,” then they could face the exact same legal and reputational exposures as Amazon.
To complicate matters still further, maritime tech vendors face compliance obligations across multiple jurisdictions. Training AI on crew biometrics or voice streams violates strict GDPR provisions regarding explicit consent and purpose limitation. Furthermore, under the Maritime Labour Convention (MLC), seafarers’ right to privacy in welfare communications is protected.
To avoid perpetual litigation and rebuild creator trust, tech providers must move away from surreptitious data collection and establish fair, transparent framework standards.
First of all, platforms must implement clear opt-in frameworks for generative AI training. If data is solely used to power creator tools, such as automated channel captions, that utility must be cleanly isolated from external, commercial AI models.
Secondly, tech providers should segment operational telemetry from personal/crew data and never train AI models on personal content without explicit licensing. In fact, explicit terms of service for shipowners and crew should be clearly defined in any contract, stating clearly whether data is processed solely to deliver the service or used to refine general AI capabilities.
Finally, independent third-party audits and robust metadata tracking are required to verify when creator content has been scraped and to guarantee that opt-out requests cleanly remove data from future training pipelines.
Without these structural shifts, tech companies risk alienating the very creator communities that give their platforms value. Sustainable innovation requires mutual consent, fair compensation, and respect for digital intellectual property.
Words by Phil Page, Head of Public Relations, Intent Communications Photo Credit: Unsplash.com
———————-
Like this article? Click here to sign up for our free monthly ‘De-Mystifying Digitalisation and AI in Maritime’ newsletter and get the latest cross-sector expert insights, news, interviews, videos and case studies straight to your inbox. You can also buy a ticket to the Demystifying AI in Maritime 2026 London Summit here.