White House launches America.gov AI chatbot for federal services

By Splitfeed EditorialLast updated Sep 29, 2026, 21:40 UTC

On September 29, 2026, the Trump administration launched America.gov, an AI-powered website and chatbot designed to help Americans find and navigate federal government services. Trump signed an executive order on the same day formally integrating federal services with the platform, and he spoke at the launch event, declaring he would never stifle the growth of technology. The site uses a language model to answer questions by drawing on government sources.

The launch drew immediate attention for an awkward reason: the chatbot's own answers contradicted statements Trump made during his remarks promoting it. According to PBS News, the site's question-and-answer function confirmed that Trump's name appears in the Epstein files and that Joe Biden won the 2020 election, both claims that cut against positions Trump has taken publicly. The White House has not addressed the discrepancy. Separately, TechCrunch flagged that language models remain prone to hallucinations, a known limitation that raises questions about whether a chatbot handling official government information can be relied upon by the public.

The pro-AI camp has not yet published reactions to this story. Given the nature of the launch, that camp would typically argue that AI-assisted access to government services reduces friction for ordinary citizens and that reliability concerns are solvable engineering problems, but no one from that camp has said so on the record here. The anti-AI camp has been vocal, with concerns running in two directions: accuracy and privacy. Users including Daily Kos and the account kdjs-1.bsky.social flagged that the associated app reportedly tracks location and contains a keylogger, and the account crawf34.bsky.social suggested the administration would try to make the AI produce different answers going forward. The middle ground, represented by Bloomberg Government and the account Tuffy, frames the launch as an unsurprising step in a broader government AI trend, noting that some version of this would likely have happened regardless of which administration was in power, even as public skepticism about AI's economic and social effects is growing.

The most immediate things to watch are whether the White House modifies or restricts the chatbot's outputs in response to the factual contradictions that surfaced at launch, whether independent security researchers confirm or refute the tracking and keylogger claims about the app, and whether the executive order faces any legislative pushback, a thread the New York Times appeared to be pulling on in its concurrent reporting about GOP lawmakers and data center expansion pressures.