Cover image

Who Acts, Who Decides: Four Patterns for Human–AI Work

TL;DR for operators AI autonomy and AI initiative are different design decisions. An AI system can execute work independently while waiting for instructions, or proactively offer help while remaining limited to a narrow task. Michael Weiss identifies four collaboration patterns: Instruction, Delegation, Assistance, and Co-Creation. Each suits a different kind of work, and each introduces a recurring tension. Delegation can increase capacity while reducing oversight; Assistance can improve individual components while weakening their fit with the whole task. ...

October 23, 2026 · 9 min · Zelina
Cover image

A Valid Credential Can Still Identify the Wrong Agent

TL;DR for operators An AI agent can authenticate successfully while its credential, software client, or external identity record is associated with the wrong persistent actor. This becomes consequential when agents operate across deployments, rotate credentials, and accumulate audit histories. SoulAuth proposes a stable ActorIdentity for each independently identifiable Human or long-lived AI actor. Credentials, clients, sessions, and cross-domain identity relationships remain separate objects with their own lifecycles. Authentication proves an identity under specified conditions; it does not automatically establish permission to act. ...

October 22, 2026 · 9 min · Zelina
Cover image

A CAPTCHA Pass Is Not a Human Response

TL;DR for operators Low-cost autonomous agents can complete paid online surveys without recurring model-access fees, and conventional bot screening does not reliably identify them. In a controlled comparison of nine agent configurations, three fully open systems completed the survey, never triggered paste detection, and consistently failed a particular hidden-instruction check. Commercial agents exhibited different detection failures.1 ...

October 21, 2026 · 9 min · Zelina
Cover image

Every Tool Call Succeeded. The Workflow Still Failed.

TL;DR for operators An AI agent can complete every reported tool call while leaving a business workflow in an inconsistent state. The critical distinction is between successful invocations and the external effects that remain after the workflow commits or aborts. A study of 98,291 remotely discovered Model Context Protocol (MCP) tools found that 74.0% emitted at least one standard annotation field. Yet the standard annotation vocabulary could not fully express any of eight transactional capabilities identified by the researchers. ...

October 21, 2026 · 8 min · Zelina
Cover image

Changing One Number Is Not Yet Forging the Document

TL;DR for operators When a lender, insurer, auditor, KYC provider, or tax authority receives a PDF as evidence, what exactly makes that document trustworthy? Changing the requested value is only one layer of success. In this benchmark, 81.1% of 1,750 edits passed the base verifier, but only 46.2% also survived checks that the change was visible, correctly localized, typographically plausible, and consistent throughout the document. Among 1,373 decidable successful edits, 539 still contained the original value elsewhere. ...

October 11, 2026 · 8 min · Zelina
Cover image

More Proof Machinery, More Compute: What TCSAlgBench Says About Research Agents

TL;DR for operators When a research agent fails on a difficult proof, the next question is how much more inference to spend: another independent attempt, verifier feedback, decomposition into intermediate claims, search across alternative proof branches, or a larger revisable plan. TCSAlgBench makes that allocation problem measurable on research-level theoretical computer science. ...

October 7, 2026 · 8 min · Zelina
Cover image

Who Registered This Agent? ATI Builds Trust Into Discovery

TL;DR for operators When hundreds of agents span internal teams, vendors, and partner domains, discovery has to answer more than which agent can do the task. Operators also need to know who registered it and what evidence justifies trusting its identity. Agent Trust Infrastructure (ATI) addresses those questions by separating trusted-registry governance, per-agent publication, and resolver-side discovery rather than collapsing them into one directory.1 ...

October 7, 2026 · 6 min · Zelina
Cover image

A Green Check Is Not a Physical Verdict: What SimVerity Changes About Agent Deployment

TL;DR for operators A simulator can clear an agent while the physical deployment simultaneously fails one operational claim and satisfies another. In the paper’s 240-trial calibration block, every source-cleared execution was a false clearance for immediate completion, while only 20 of 240 failed on reported state, 42 of 240 failed on observable physical effect, and none failed on the eventual settled postcondition. ...

September 23, 2026 · 7 min · Zelina
Cover image

Memory Has a Budget: Compress Long-Running Context by Value, Not Age

TL;DR for operators Fofadiya and Tiwari propose a context-management system for long-running LLM interactions that does more than summarize old conversations.1 It scores historical turns by relevance, recency, dialogue dependency, and coherence, then applies a three-level policy: retain high-value turns verbatim, summarize medium-value turns, and remove low-value history when the context budget requires it. ...

September 19, 2026 · 6 min · Zelina
Cover image

The Task Is Larger Than the Prompt: What Agents Miss Before They Act

TL;DR for operators A tool-using agent can complete the action a user requested and still produce the wrong operational outcome. The missing requirement may be recoverable from current device state, a temporary preference, an accessibility setting, a privacy boundary, or the reversibility of the requested change. Implicit Intelligence – Evaluating Agents on What Users Don’t Say1 tests this problem directly. Across 205 deliberately challenging scenarios, the strongest evaluated model, GPT-5.2-pro, achieves a 48.3% Scenario Pass Rate: fewer than half of scenarios satisfy every required criterion. Its mean Normalized Scenario Score is higher, at 72.7%, showing that agents often complete substantial parts of the task while still missing at least one consequential requirement. ...

September 8, 2026 · 8 min · Zelina