If we all start to reject our posts used for training models, will it reduce the influence of our unique voice and views on those models? Increasingly, the models will be the main window of everyone in the rest of the world. It seems that people who are interested in these things will be the people with the most data to end the default behavior training of the models.
The influence of others
To be honest, it was uncomfortable for me that the internet users were forced to refuse Artificial intelligence Training as default. It would not be good if the agreement asserted as a standard for generation AI companies for them Scrape the web And any other data warehouse they can find to build bigger and bigger border models?
But, unfortunately, it is not the case. The companies like Openai And Google argue that if Reasonable use For all these data has been taken from them, none of these will even be possible. Currently, users do not want to contribute to the general models trapped with some refusal processes on different websites and social media platforms.
Even if the current bubble is around the generation Dotcom bubble After a few years, the models that provide energy for all the new AI tools will not be extinct. Therefore, the ghosts of the post on your appropriate forum and the topics of social media support the strongly organized judgments that will live inside the software tools. You are right when choosing, which means positively trying not to be included in a part of a long -lasting culture.
To solve your question directly and reality, these things The refusal processes It is basically useless in their current state. Those who refuse right now are still affecting the model. Suppose you fill out a form for a social media website so as not to use or sell your data to train AI. Even if that platform respects that requirement, there are countless Silicon Valley startups with 19 -year -old Plucky, who will not think about scanning data on that platform, even if they are not technical. According to the general rule, you may think that anything you've ever posted online can turn it into many general models.
Ok, but assuming you can actually block your data from these systems or ask it to be deleted after reality, will make your voice or impact on your voice or impact on AI tool? I thought about this question for a few days, and I was still torn.
On the one hand, your small information is only a small contribution to the broadness of the data set, so your voice, as a character or author is not public, is likely not to affect the model in one way or another.
From this view, your data is only another brick in the wall of a 1,000 -storey building. And it is memorable that data collection is only the first step in creating AI model. Researchers spend months tweaking the software to get the results they want, sometimes relying on Low salary labor To label the data set and evaluate the output quality to screen. These steps can continue abstract data and reduce your personal impact.
On the opposite side, what if we compare this with voting in an election? Millions of votes were chosen in US presidential elections, but most citizens and democrats were asserted that all votes were important to each other, with a continuous restraint of making your voice heard. It is not a perfect metaphor, but what if we think our data has the same effect? A small whisper between the sound of the noise, but still affects the output of the AI model.
I was not completely convinced of this argument, but I didn't think this view was completely rejected. Especially for experts on topics, different insights and approach your information is the only value for AI researchers. Meta will not experience the trouble of use All of those books In its new AI model if any old data will perform the procedure.
Towards the future, the real impact that your data may have for these models may be inspired by inspiration. Synthetic data of the Viking people. When companies create general AI systems out of quality information to eliminate, they will enter their Ouroboros era; They will start using AI generation to copy human data and then they will return to the system to train the next AI model to better copy human reactions. As long as the general exists, just remember that you, as a human being, will always be a small part of the machine, whether you want or not.