Spawning AI’s New Platform to Empower Artists in AI Model Training

Creating Autonomy for Artists in the Digital Realm

In an era characterized by rapid advancements in artificial intelligence, Spawning AI co-founders Jordan Meyer and Mathew Dryhurst have unveiled Source.Plus, a novel platform dedicated to redefining the dynamics of digital content utilization. This project distinctively focuses on generating a meticulously curated and ethically-sourced data set that guarantees artists retain control over the distribution of their creations.

This initiative has birthed a data collection consisting of approximately 40 million images, sourced predominantly from the public domain and under Creative Commons’ most liberal CC0 license. The founders of Spawning AI emphasize that although their database may be smaller in size compared to rival AI training sets, it does not compromise on quality, proving sufficient to train cutting-edge image generation models.

Revolutionizing Rights Management and Data Integrity

Source.Plus emerges as a game-changer amidst ongoing discussions on the ethical considerations surrounding generative AI practices. The platform aims to establish a seamless and transparent system that empowers creators to submit their works specifically for generative AI model training, essentially creating an ‘opt-in’ model for rights holders.

The founders have expressed that conventional AI training often resorts to the path of least resistance, employing readily available yet potentially contentious data. This practice has led to artists and rights holders feeling disenfranchised, lacking agency over how their works are appropriated for AI training purposes. Source.Plus represents a conscious effort to shift this paradigm towards a model that respects and safeguards creators’ rights.

From Concept to Reality: Source.Plus Enters Limited Beta

Building upon Spawning AI’s foundational tools for tracking art provenance and managing usage rights, Source.Plus has already launched in a limited beta version. The platform is not just a static repository; it positions itself as an enrichment platform packed with a repertoire of features designed to nurture the entire training pipeline.

The initiative does not stop at compiling a free-to-use CC0 data set but also endeavors to establish an elevated standard for fair data sourcing, distinguishing itself from other companies that often grapple with ensuring ethically sourced data. Meyer has indicated a stark commitment to transparency, including rigorous validation of image licenses to protect against copyright infringements, while maintaining a stance against contentious content by employing classifiers and human moderators.

Advancing Fair Compensation

Addressing the pitfalls faced in existing compensation models for artists’ contributions to generative AI, Source.Plus envisions a more equitable approach. As the beta phase concludes and the platform expands its data set offerings, it will present a model that champions the creators’ ability to dictate prices and conditions for their works, a marked departure from industry standards that often overlook the creators’ financial interests.

Key questions and answers:

What is Source.Plus?
Source.Plus is a platform created by Spawning AI aimed at transforming how digital content is used for AI model training. It allows artists to maintain control over their works, offering an ethically-sourced data set that contributors can opt into.

How is Source.Plus different from other AI data sets?
Unlike other data sets that may use content without clear consent from the rights holders, Source.Plus provides a collection sourced primarily from public domain and CC0-licensed images, and offers an ‘opt-in’ model for artists, ensuring respect for their rights.

What challenges does Source.Plus address?
It tackles the ethical concerns of how data for generative AI is sourced, rights management, and the need to provide fair compensation to the artists whose works are used for training AI models.

Advantages:

  • Ethical Data Sourcing: Ensures that artists who contribute content to the data set have agreed to its use, thereby addressing copyright issues.
  • Artist Autonomy: Gives creators the power to define how their work is used and under what conditions, including compensation.
  • Quality Over Quantity: Focuses on a smaller but high-quality data set, reinforcing the importance of data integrity.
  • Transparency: A commitment to transparency regarding image licensing and content moderation.

Disadvantages:

  • Limited Size: The data set size may be smaller compared to competitors, which could potentially limit the diversity and the breadth of AI model training.
  • Market Acceptance: As a new model within the AI industry, Source.Plus may face challenges related to user adoption and recognition.

Main domain related links:
To learn more about the broader context of how artificial intelligence interacts with digital rights management and artist compensation, visit:

Creative Commons – to understand the licenses that enable creators to mark their works with the freedoms they want it to carry.

Key challenges or controversies:
Ensuring the inclusion of diverse content while respecting artists’ rights remains a balancing act. Additionally, the AI industry’s widespread practice of using data without proper attribution has sparked debates about privacy, copyright, and ethical AI use – issues Source.Plus aims to mitigate.

Controversies:
There’s an ongoing discourse on the nature of consent and remuneration in the utilization of artworks for training AI. Pushback from the art community against tech companies has been palpable, shaping the conversation around the ethical sourcing of data. Source.Plus is a direct response to these controversies, aspiring to set a precedent for responsible data use and artist empowerment.

Privacy policy
Contact