alexgarden

5 ms·

31.
by alexgarden 8mo ago
If we've arrived at the point where bots are impersonating me (instead of the billions of other choices), I'm probably at peak Alex. I'll light a candle. So... easy to disambiguate this one. I got the idea for Homeworld one
32.
by alexgarden 8mo ago
Go Canada! I personally can't wait to see what happens to the world when all of us find the passion to create again.
33.
by alexgarden 8mo ago
Wow... I really relate to this. I'm 50 as well, and I started coding in 1985 when I was 10... I remember literally every evolutionary leap forward and my experience with this change has been a bit different. Steve Yegge recently did a
34.
by alexgarden 8mo ago
Very silver-lining viewpoint. I suppose that when tens of thousands of users have their identities leaked by well-intentioned helper agents, we will certainly elevate the security discourse. ;) My personal opinion is that transformer arch
35.
by alexgarden 8mo ago
Hunter S. Clawmpson is an OpenClaw agent wrapped with smoltbot (github.com/mnemom/smoltbot), which implements the Agent Alignment Protocol (github.com/mnemom/aap) — a transparency layer for the agent stack. Every decisio
36.
The Agent Internet Is Building Its Immune System in Public(mnemom.ai)
3 points by alexgarden 8mo ago|1 comments
37.
by alexgarden 8mo ago
I like your analogy of OpenClaw==Linux. Paradoxically, Linux was simple and secure in its early days due to (among other reasons) being open source from (almost) the beginning. Vulnerabilities were found and fixed by the community. Ope
38.
by alexgarden 8mo ago
This will remain an issue as long as we rely on the agent substrate's good judgment to power OpenClaw (or whatever comes next). To make these frontier-based agents performant, they're weighed heavily on "bias to action"
39.
by alexgarden 8mo ago
I think the issue is that you're framing alignment in the context of an individual action, "Are we aligned in this task?", which, I agree, devolves into a rules-based approach. I think the notion of Alignment is actually the
40.
by alexgarden 8mo ago
The truth is that the ship on "rules-based systems" has sailed. Doesn't matter if the vector is prompt injection, malicious payloads in skills, or backdoors - your agent (you will end up with one) is going to be exposed to j