To avoid lock in, you need the ability for the user to credibly exit.
Exit Chat Control · Become Ungovernable
A practical field guide to escape Chat Control and reclaim your digital privacy: encrypted messaging, email, VPN, DNS, passwords, Linux, GrapheneOS, self-hosting. Free, open source, no tracking.
Introducing Automatic Key Verification
Signal now offers a feature called “automatic key verification” which complements the existing safety number system. Signal is always end-to-end encrypted, and automatic key verification provides an additional, streamlined way to confirm that there’s no unexpected party between you and the other ...
OpenAI unveils Lockdown Mode to protect sensitive data from prompt injection attacks | TechCrunch
Even with Lockdown Mode, ChatGPT could be vulnerable to prompt injections, but the goal is to reduce the likelihood that sensitive data gets shared in the process.
Trust & Safety Library - Trust & Safety Professional Association
Welcome to the Trust & Safety Library! We use this space to collect articles, blog posts, journal articles, lectures, podcasts, and websites that trust and safety professionals may find useful in developing policies, supporting moderators, building systems to detect violations, and generally deepening their practice. We welcome your submissions and feedback. This project was initially
DropLock
The link below is like an open lock box that belongs to you. Anyone you share it with can put a secret inside and create a locked link that only you can open.
The Session You Cannot Take With You | EARENDIL
Inference APIs are filling sessions with encrypted reasoning, hidden search results, opaque compaction, and encrypted subagent messages. A growing form of lock-in.
https://www.quera.com/blog-posts/holding-the-light-teaching-an-ai-to-lock-and-tune-our-quantum-computers-lasers
QuEra's quantum computers run on lasers held at exact frequencies, a precision no eye can see and only the atoms can distinguish. Keeping a laser there is a continuous act: temperature, vibration, and pressure push it off target all day, and layered feedback pushes it back. When the feedback loop fails, the lock breaks and the machine stops. Bringing it back has historically taken one specific expert: someone who knows the lasers intuitively and has experience with the exact recovery sequence required to return it to the right state. If the lock broke in the middle of the night, that person had to drive to the lab and fix it.
The agent control plane gets real - Sensemaker
Two prompt-injection incidents show why agent security is about permission boundaries, not better instructions.
Wondering if using "credible exit" when talking about atproto will confuse/mislead people how atproto works under the hood?
𝕮
This is a radically different idea than the advocacy I’ve seen for “data portability” and ”data sovereignty”. It’s a *restructuring* of power rather than small adjustments to existing structures. atproto and local-first, both in their own ways, demonstrate what such a restructuring can look like.
I signed up for science with verify later. You can only "explore W" while not verified (no account changes or posting). I then tried to sign in on Mu, just to see what would happen, and it wants WID verification via QR. So it looks like the account is locked behind verification, even off platform.
don't think i'm worried about this in particular, but it does make me wonder what exactly is the credible exit for my own data if a space authority becomes malicious and stops issuing me credentials? (still wrapping my head around the permissioned data proposal so sorry if this is a dumb question!)
Andrew Lisowski 💻
I really worry that permissioned data is gonna kill a lot of stuff Take my recipe app: If i wanted no one to have access to the records but my app all i have to do is 1. create a space 2. use that space for all user content Bam, now no open data
