Inbred, 'Habsburg AI' is coming.
Though whether it sticks in the long term remains to be seen, at least for the time being generative AI seems to be cementing its place in our digitalreal lives. And as it becomes increasingly ubiquitous, so does the synthetic content it produces. But in an ironic twist, those same synthetic outputs
To understand this phenomenon better, we spoke to machine learning researchers Sina Alemohammad and Josue Casco-Rodriguez, both PhD students in Rice University's Electrical and Computer Engineering department, and their supervising professor, Richard G. Baraniuk.
And the crux is that increasingly, AI models are being trained on not just natural data, or real data sourced from the real world. Now, they're also being trained with data that's been synthesized by other generative models. So consequently, we've entered an age where either wittingly or unwittingly, generative models are increasingly consuming the outputs from other generative models.
Let's look at this intuitively. Say you have 1 billion pieces of natural data, and you have one piece of synthetic data. MAD won't happen. But one year later, if you have 1 billion pieces of synthetic data, then definitely it'll go MAD in five iterations. We have found this ratio in Gaussian modeling.Right. And just to be clear, there absolutely is a threshold for every model.
That's the thing that's really the most dangerous, and you might not even realize it's happening. And by drift away from reality, I mean you're generating images that are going to become increasingly, like, monotonous and dull. The same thing will happen for text as well if you do this — the diversity of the generated images is going to steadily go down.
Canada Latest News, Canada Headlines
Similar News:You can also read news stories similar to this one that we have collected from other news sources.
Bavaria's Data Protection Office Launches Probe Into Worldcoin's Iris-Scanning Data Practices – Bitcoin NewsBavaria's State Office for Data Protection Supervision (DPS) has initiated an inquiry into the Worldcoin project and its data collection techniques. WLD
Read more »
Instagram seems to be working on labels for posts 'generated by Meta AI' | EngadgetMeta appears to be working on several new generative AI features for Instagram, including labels that allow creators to identify images “generated by Meta AI.”.
Read more »
19 Celeb Couples Who Seem Like They Were Randomly Generated Together With AISome are unexpected but sweet, and others may send shivers down your spine.
Read more »
YouTube tests AI-generated video summaries | EngadgetYouTube is testing AI-produced video summaries that can describe clips before you watch them..
Read more »
We Talked to The Researchers Studying How AI-Generated Content Breaks AI's BrainWe talked to Rice University scientists studying how training AI with synthetic, AI-generated content causes a destructive inbreeding effect.
Read more »
Instagram labels would identify Meta AI-generated images: reportMeta Platforms is reportedly creating labels that allow Instagram users to identify images that are 'generated by Meta AI.' It comes after Meta agreed to adopt AI safety measures.
Read more »