alt Hacker NewsI don't understand why you would want to remove well-balanced and relatively small local-inference models that ship with every mac.
Sure, not frontier levels but fine for basic tasks and they are off-the-cloud.
Replies
Because the Macbook Air still comes with only 512 GB of SSD in its default config. A feature that you rarely use probably isn't worth even 1/50th of your meager storage.
The point is that it downloads a multi-gig model automatically without providing an easy way to reclaim that storage space if you don't intend to use that feature.
This is very much an anti-user behavior. Certainly having models run on-device is preferred when possible, but if you don't want to use the Apple Intelligence this is an anti-feature. Especially on devices with small storage capacities which are impacted the most by this.
Even if you don't want to personally remove them, it should be easy and front and centre to do so.
It's my disk space, processing power and electricity. Apple shouldn't get to use it as they please
Here are a few reasons someone would want that. You might not value them the same, but that does not make them wrong.
1) They have issues with the ways these models were trained. They may consider it unethical to use models trained on what they consider plagiarized material or on material that used a lot of energy for training.
2) They may have higher value uses of the storage space and compute resource.
3) They may view installation of apps and features without consent to be an infringement on running their device the way they would like.
4) They may find the general technology and its impact on society to be concerning, anxiety inducing, frustrating, or irritating. And simply not want to participate.
5) They think it sucks. Like, out just doesn't work or doesn't help them.