Member-only story
https://www.wired.com/story/stack-overflow-will-charge-ai-giants-for-training-data/
Here’s a lesson on how dreams die. Use open source to craft an engine and watch refiners cut the gas supply. This movement will do a few things.
Create a class system for supply and demand
Enhance bias
Drive self referential data models that flatten diversity within training sets
Drive investors to force more data paywalls while simultaneously demanding startups use more AI in creation.
Don’t trust me. Let’s hear from the affected party. Chatgpt
“When individual data sources are locked behind paywalls, it can have a significant impact on the training and performance of AI systems that rely on that data.
AI systems typically require large amounts of high-quality data to learn and improve their performance. If essential data sources are not accessible due to paywalls, the AI system may not have access to the necessary information to develop accurate models.
Depending on the nature of the paywall, there may be ways to obtain the data legally, such as by purchasing a subscription or negotiating a data-sharing agreement. However, this may not always be feasible or financially viable for the developers of the AI system.
In some cases, developers may resort to using alternative data sources that are free or publicly available, but these may not be as comprehensive or representative of the target population, which could lead to biased or inaccurate results.
Overall, the unavailability of critical data sources due to paywalls can limit the performance and effectiveness of AI systems, and may ultimately hinder progress in the field of AI.”
