hierophantos
I stumbled into two recent papers, thinking that it was about "capabilities" in the OCap sense and datalog (a related technology to what I'm working on): • https://arxiv.org/html/2603.26725 Capability Safety as Datalog: A Foundational Equivalence; Cosimo Spera; March 2026 • https://arxiv.org/html/2603.15973 Safety is Non-Compositional: A Formal Framework for Capability-Based AI Systems; Cosimo Spera; March 2026
Since capabilities in the OCap sense are just transitive closure of references on some reference graphs, it turns out that these two concerns with conflated terminology are nonetheless the same.
Curious if you or anyone has thoughts on managing capability escalations in multi-agent systems. Some thoughts be a-brewing...
Talking to one of the "O-Cap Gang", Norm Hardy (http://www.cap-lore.com/CapTheory/) a few weeks ago, he said, to the point of creating secure distributed computing architectures: "Object capabilities are not easy, but they're the only way I see forward."
I'm biased to agree.
That you are posting this, shows that he's got someone in his corner.
Bees are awesome, by the way. I'd invest. :)