Google Has Banned the Training of Deepfakes in Colab

Sometime in the last two weeks, Google has quietly changed the terms of service for its Colab users, adding a stipulation that Colab services may no longer be used to train deepfakes. ...

Of the two popular deepfake-creation distributions, DeepFaceLab (DFL) and FaceSwap, both of which are forks of the controversial and anonymous code posted to Reddit in 2017, only the more notorious DFL appears to have been directly targeted by the ban. According to deepfake developer ‘chervonij’ at the DFL Discord, running the software in Google Colab now produces a warning:

‘You may be executing code that is disallowed, and this may restrict your ability to use Colab in the future. Please note the prohibited actions specified in our FAQ.’

However, interestingly, the user is currently allowed to continue with the execution of the code. ...

Since deepfake training is a VRAM-hungry pursuit, and since the advent of the GPU famine, many deepfakers in recent years have eschewed home training in favor of remote training in Colab, where it’s possible, depending on chance and tier, to train a deepfake model on powerful cards such as the Tesla T4 (16GB VRAM, currently around $2k USD), the V100(32GB VRAM, around $4k USD), and the mighty A100 (80GB VRAM, MSRP of $32,097.00), among others.

The ban on Colab training seems likely to reduce the pool of deepfakers able to train higher-resolution models, where the input and output images are larger, more suited to high-resolution results, and capable of extracting and reproducing greater facial detail. ...

See the full story here: https://www.unite.ai/google-has-banned-the-training-of-deepfakes-in-colab/