What you gonna do, Luke Skywalker?

A follow-up to [this note] given Anthropic's recent announcement that they're going to require government-issued ID to use "certain features" of their models. They're using none other than Peter Thiel-backed Persona for this. Ugh.

Being responsible with powerful technology starts with knowing who is using it. Identity verification helps us prevent abuse, enforce our usage policies, and comply with legal obligations.

We are rolling out identity verification for a few use cases, and you might see a verification prompt when accessing certain capabilities, as part of our routine platform integrity checks, or other safety and compliance measures.

We only use your verification data to confirm who you are and not for any other purposes.

So, is this the new form of 'export controls' in a hyper-connected world? You're allowed to use these tools, but only if you're the right nationality (or race, or religion?) Given the ultra-nationalist, if not fascistic, US administration, they are likely to at least try to mine everything you put into a US-owned Big Tech AI tool in the "interests of national security".

We have, therefore, three options as far as I see it:

1. Comply and attempt to encrypt or obfuscate everything before it goes off to to Big Tech servers. This would be awkward, unrealistic, and still means that you have to go through the ID process.
2. Run everything locally. Buy the hardware to be able to own your entire inference stack so that nothing leaves the machines you own. Costly.
3. Co-operate on AI compute using something like co/core which is brand new and somewhat unproven. Potential privacy issues.

Interesting times.