Stormy Whether
GDPR services is a market we shouldn't need at all, but it's growing like a thundercloud. (Source: Mordor Intelligence)
Toward Agentic VRM
vCon is "A new IETF open standard defining how conversation data — audio, video, text, and metadata — is structured, signed, and exchanged across any system. The foundation of VCONify." More:
A vCon (Virtualized Conversation) is a standardized JSON container for all data related to a real-time human conversation. Think of it as the universal format for conversations — the way PDF standardized documents or MIME standardized email.
One vCon carries everything: the audio recording, video stream, text transcript, participant identity, timestamps, derived analytics, and a cryptographic signature that proves the record hasn't been altered.
It is platform-agnostic, vendor-neutral, and AI-ready by design.
A Brake on the Breaks
ReverseAlignment says,
Most of the effort and capital in AI goes to two tasks: making systems more capable, and aligning those systems with human values.
Both matter. But even a perfectly aligned system will underdeliver, or do real harm, if it is released into economies, governments, and information systems unprepared to receive it.
We call that neglected work reverse alignment: the deliberate redesign of institutions, norms, skills, and governance so that societies can absorb AI safely and share its gains widely.
They have a coalition.
Along those same aligns,,,
Functional Consciousness is the site for a paper: Functional Consciousness: A Proxy Metric Using Self-Models, about which the abstract says,
This paper proposes Functional Consciousness (FC) as a measurable architectural property: the observable capacity of a system to access and reason about internal representations of its own states. We introduce a computationally tractable metric on FC that operationalizes core tenets of major consciousness theories through self-models and their associated reasoning power, measured through informational richness and state-space expansion under inference. The resulting Functional Consciousness Score (FCS) is applied to benchmark systems with known internal structure, including a Waymo L4 autonomous vehicle. To extend the framework to black-box systems, we present Functional Self-Model Analysis (FSMA), an abductive methodology for inferring self-models from behavioral evidence. Applied to stream-of-consciousness literature, FSMA yields a catalog of self-models that serves as a reference for estimating functional consciousness in more complex biological and artificial agents. The resulting scores align with intuitive gradients of cognitive sophistication while remaining operationally grounded. Finally, we compare FC with major theories of consciousness and argue that several of their central functional claims become partially measurable within this framework.
Frontiers of Importance
[Agents of Chaos](Agents of Chaos) is an academic study abstracted this way:
We report an exploratory red-teaming study of autonomous language-model-powered agents deployed in a live laboratory environment with persistent memory, email accounts, Discord access, file systems, and shell execution. Over a two-week period, twenty AI researchers interacted with the agents under benign and adversarial conditions. Focusing on failures emerging from the integration of language models with autonomy, tool use, and multi-party communication, we document eleven representative case studies. Observed behaviors include unauthorized compliance with non-owners, disclosure of sensitive information, execution of destructive system-level actions, denial-of-service conditions, uncontrolled resource consumption, identity spoofing vulnerabilities, cross-agent propagation of unsafe practices, and partial system takeover. In several cases, agents reported task completion while the underlying system state contradicted those reports. We also report on some of the failed attempts. Our findings establish the existence of security-, privacy-, and governance-relevant vulnerabilities in realistic deployment settings. These behaviors raise unresolved questions regarding accountability, delegated authority, and responsibility for downstream harms, and warrant urgent attention from legal scholars, policymakers, and researchers across disciplines. This report serves as an initial empirical contribution to that broader conversation.
The list of co-authors is half as long as that paragraph.
So I'll wait
I want a Mac Studio for M5-grade personal AI and 16 TB of storage for my massive photo file collection. Wired calls it "only a preview of what’s to come." Only $18,299.00.
So I'm not
The Observer says, "AI giants are trying to sell us the apocalypse. Don’t buy it." Subhead: "Corporate spiel about the dangers of the technology is paradoxically just a way of marketing it as a transformative power of world-changing significance."
And much else that makes good and useful reading
In The death of history's most successful publisher, Cory says "Google has a scorching case of 'main character syndrome.'"
Doesn't pay to get old
In How the Government Forces Doctors to Defraud Our Seniors, Matt Bivens M.D. says, "Medicare has decided it can avoid paying its bills by simply changing the meanings of basic words. ER doctors like me sign off on this every day."
Leave a Reply