Anthropic PR: "Ma'am, you opted IN to training on your therapy sessions and intellectual property and algorithms and salary and family history!" Don't you remember the modal???
The Modal: https://imgur.com/afqMi0Z
111–120 of 235 posts
Anthropic PR: "Ma'am, you opted IN to training on your therapy sessions and intellectual property and algorithms and salary and family history!" Don't you remember the modal???
The Modal: https://imgur.com/afqMi0Z
I don't know what they've been training on but I just canceled claude for the second time. Besides the numerous UI bugs of the web interface, incessant flickerings, it has gotten weirdly super condescending and negative in a way I hadn't observed neither in the past nor with other llms. Probably that people accused it of being sycophantic and they have tried to adjust it but they didn't do it well. It'd rather critic…
Title is misleading, they're now opt-out rather than opt-in to your data being used for training. All you have to do is flip a single switch in the options to turn it off, I don't understand why everyone is treating this as being such a big deal. Edit: I just logged in to opt out, they presented me with the switch directly. It was two clicks.
I still expect that our conversations will not leave the premises (ie end up on the internet), because that would be something else, but other than that, I knew what I signed up for.
This going to turn into one of those situations where we find out they trained on everyone whether they opted-out or not down the line. I want to keep using Claude, but I also don't want all the solutions I come up with to become common knowledge.
Am I the only one that assumed everything was already being used for training?
This going to turn into one of those situations where we find out they trained on everyone whether they opted-out or not down the line. I want to keep using Claude, but I also don't want all the solutions I come up with to become common knowledge.
Modern AI is built on data. Trusting that our conversations will not be used for training the model is a bit like giving a glutton their favourite food but making them promise not to eat it. Sure, why not. I still expect that our conversations will not leave the premises (ie end up on the internet), because that would be something else, but other than that, I knew what I signed up for.
To be honest, these companies already stole terabytes of data and don't even disclose their dataset, so you have to assume they'll steal and train at anything you throw at them