How the answers frame USPS
This snapshot covers 250 benchmark questions put to ChatGPT, Claude, Gemini, Perplexity and Grok on 2026-09-29. None of the questions named a brand. USPS was named in 589 of 1,250 answers. That puts it third among logistics brands, behind FedEx (898) and UPS (885) and ahead of DHL (484). By provider, Grok named USPS in 155 of 250 answers and ChatGPT in 141. Perplexity named it in 74 and Claude in 96.
The answers place USPS in a specific role. Visibility was highest on questions about package handling (76%), value (74%) and small-business fit (74%). It was lowest on innovation (22%), leadership (22%) and sustainability (23%). The credited themes follow the same pattern: Price & Value and Reliability & Operations each drew six credits, and Customer Experience drew two.
The price framing was explicit. In Claude's answer on shipping packages under 2 pounds, it said "USPS Ground Advantage is the cheapest way to ship for most small, light packages under about 2 lbs". Perplexity, asked which carriers small-town shop owners recommend, said "USPS is often a practical pick for lightweight parcels and rural destinations". ChatGPT's answer on mail-order medications for an elderly parent pointed to USPS as the practical choice for routine shipments and cited its published on-time figures. It sent time-critical shipments to UPS.
Where the question turned to tracking or claims, USPS came out weaker in several answers. ChatGPT's tracking answer said USPS "tends to lack detail." Claude, asked about porch theft, put USPS under "More Challenging Carriers". Claude's tracking answer said packages "can go missing in the Postal Service's network for weeks or even months with no tracking updates". Perplexity's ETA answer put USPS's on-time rate below UPS and FedEx. Gemini and Grok pointed to USPS for its unionized workforce and for legal documents sent by Certified Mail.
USPS Ground Advantage is the cheapest way to ship for most small, light packages under about 2 lbs…
What the press is saying
PeakMetrics' news index recorded 28,258 mentions of USPS over the last 30 days. In 6,018 of them the brand was in the headline. That is far more coverage than DHL (10,710) or Amazon Logistics (177), and in the same range as FedEx (21,769). UPS is higher at 34,392.
The recent headlines are mostly about elections. They cover a whistleblower raising concerns about what one outlet called untested USPS election technology, an attorney discussing claims about the mail-in voting system, and a former USPS worker indicted for allegedly discarding mail-in ballots. The one seasonal item is USPS Operation Santa accepting Christmas letters and seeking adopters.
USPS stolen packages and missing mail can be more challenging than packages lost by other carriers because it's trickier to file a claim with USPS.
The gap
The two stories barely overlap. The press is reporting on mail-in voting, ballots and an election-technology whistleblower. The answers are about postage rates, drop-off convenience, small-seller shipping and how well tracking works. None of the supplied answers touched on election or ballot themes.
Part of that follows from scope. The benchmark questions covered parcel shipping, not civic or political topics, so the absence says nothing about how the assistants would answer a question about election mail. It does show that heavy press attention does not by itself change how the assistants describe USPS in delivery answers.
USPS is often a practical pick for lightweight parcels and rural destinations…
What to watch
For a USPS comms team, the snapshot points to two things. The price and small-business framing is well established across the assistants, and the weakest points are tracking detail and stolen-package claims. Those are the areas where the answers cited weaker performance, and the answers and their cited domains are the place to see which sources feed that framing.
Election-related coverage is a separate exposure. It is large in the news index and absent from these delivery answers, so it would need its own benchmark questions to see how each assistant handles it. This is a single-day snapshot of live answers, which vary between runs, so treat the figures as a baseline rather than a ranking.
