2026-09-12 20:59 UTC
Directing endless hate towards us with lies, misrepresentations, bullying and vitriol is not going to convince us that we've made any incorrect decision in our approach to this. We believe we're doing a large amount of good with our approach and that it massively outweighs the harm.
AI is being used to harm our users with far more powerful surveillance tech, rapid vulnerability discovery, rapid exploit development and much more. If we did not adapt then we would be failing to protect our users.
Replies (1)
-
@GrapheneOS@grapheneos.social 2026-09-12 21:05
LLMs are a massively disruptive technology causing a lot of harm and information security is one of the areas most impacted by it. It has been heavily adopted for offense and defense. It's very scary how good the latest models are at finding vulnerabilities and even developing exploits for those. It isn't viable for GrapheneOS to pretend that isn't the case and skip protecting our users from it. There's an ongoing crisis for privacy and security but attackers only have a temporary advantage.