When this site gets something wrong, the fix is logged here, in public, with a date. A site that hides its corrections is not more reliable, only quieter about its mistakes.
4 October 2026
The Osmium Fingerprint bench summarised Abbott et al. 2021 as "eight bipolar sulfate events between 13,023 and 12,871 yr BP". The paper's Table 1a lists eight major eruptions in those years and gives V4 deposition on Greenland and none on Antarctica, so seven of the eight are bipolar and V4 is what the paper calls a unipolar event; the complete event list the paper points to holds twelve events in the same years, ten of them bipolar. Every passage in both languages now says the eight are the table's major events and that seven of them are bipolar, and V4's card names it as Greenland only.
3 October 2026
The Nibiru bench's Sky view printed one verdict whatever the size dial was set to: the body is "about 1.5 magnitudes brighter than the Pan-STARRS Planet Nine search reached", and "There is no phase of a 3,600-year orbit at which it hides." That holds at the book's size. At Earth's size and albedo, one of the bench's own presets, the faintest the body can be today is V 22.9, 1.4 magnitudes fainter than the Pan-STARRS depth of V 21.5, and it is brighter than that depth for 1,340 of the 3,600 years. Two cards in the same view, on every phase at once and on the faintest case, carried the same fixed text, and the count of years brighter than the survey depth could run one step past the period on a short orbit (2,251 of 2,250 years). The verdict and both cards are now worked out from the dials.
1 October 2026
The David Grusch dossier said: "He said under oath that he has not personally seen a craft or a body, and that the whole of his account is what he was told by others. He describes this as the reason the claim belongs in front of a committee with subpoena power rather than in his own hands." Its opening called him "by his own description, the first whose knowledge of it is entirely secondhand." At the hearing of 26 July 2023 (transcript, p. 36), asked whether he had seen any of the bodies, he said "That is something I have not witnessed myself"; asked whether he had seen the spacecraft, he said "I have to be careful to describe what I have seen firsthand and not in this environment. But I could answer that question behind closed doors." No source was found for the sentence on a committee with subpoena power. The dossier now quotes both answers, drops that sentence, and says that most of what he has said comes, by his own description, from what others told him.
30 September 2026
The Younger Dryas Impact bench disagreed with its own arithmetic in five places. Its story said the Taurid-swarm fragment leaves "No crater, correct" and delivers "a millionth of a part per trillion"; on the bench that preset reaches the ground, leaves a crater about 830 metres wide and delivers about a thousandth of a part per trillion. It said Petaev's eight hundred metres of iron is "far too strong to break up in the air" and arrives "intact and fast"; the bench breaks it up about fifteen kilometres up, and the pieces reach the ground at almost full speed. It said that at the proponents' own ±100-year precision "The chart empties"; ten of the fifteen modelled sites drop off and five stay. It said the Hiawatha crater is "older than the ice sheet above it by a factor of four thousand" and "fifty-eight million years older than the ice sheet above it"; the bench gives no age for the ice sheet, and the factor of about 4,500 is against the boundary the crater was offered for. And it said leaving no crater "takes something a hundred times smaller"; on the bench the smallest iron that delivers the platinum is about twenty-six times wider than the largest that leaves no crater. The copy now says what the bench computes.
29 September 2026
The Kaikoura bench said its Venus preset jumps "to 0257, when Guard pointed the risen planet out to the passengers", its try-this said that at 0257 "Venus clears the sea exactly where the record says Guard pointed", and its timeline card called the light Guard pointed out "the planet Venus, risen on schedule". The bench's own sky puts Venus 4.2 degrees below the horizon at 0257, still about 1.7 degrees under the sea edge seen from 13,000 ft: held at that height it would clear near 03:07, and on the bench's modelled descent it stays under until the landing at 0315. The 0257 time comes from Maccabee's 1979 report, filed in AIR 1080/6/897, whose chronology says it was "probably during or just after this orbit" that Guard pointed out Venus, "just rising". The bench now quotes that chronology and sets the computed sky beside it, and the preset's camera faces the computed planet.
The Lazar page, the Craft topic, the Caveats page and the wiki entries on Papoose Lake, S4, Luigi Vendittelli and Bob Lazar described the image of rectangular shapes at Papoose Lake as a 2026 enhancement of 2022 Google Earth imagery; the Caveats page said: "The 2026 enhancement of 2022 Google Earth imagery shows nine rectangular shapes at Papoose Lake, which Lazar's supporters read as matching his nine-hangar description." Jesse Michels's American Alchemy episode with Lazar (2026) describes an aerial photograph taken by a pilot about 17 miles from Papoose Lake, whose contrast Luigi Vendittelli enhanced; on Chris Ramsay's DEBRIEFED (episode 83), which calls it "Luigi's image", Ramsay counts nine doors in it and Lazar says "It's exactly the way it looked." The pages now say so, and the Project Gravitour entry no longer calls the image analysis that "physically corroborates Lazar's facility description". The Tom DeLonge entry said he was largely responsible for the 2017 New York Times story "Glowing Auras and 'Black Money'", which no source in the wiki says; it now gives the story's date, two months after TTSA's founding.
26 September 2026
The Colares Incident wiki entry quoted Luis Elizondo, under a lead that names his book Imminent, as saying "A whole litany of things. A lot of the locals recall being terrified by these things, being pursued by them, almost being like a laser blast, directed energy type injury … Medical doctors came in afterwards … validating the presumption that there was some sort of directed-energy type damage." He said it in The Diary of a CEO (October 2024), not in the book, and the quotation had dropped words without marking them: he says "almost being like, like a laser blast, if you can imagine that, or directed energy type injury" and "A lot of medical doctors came in afterwards to look at the locals and the military personnel, validating the presumption". The entry now credits the interview and quotes the passage word for word, with its one cut marked.
The Kona Blue wiki entry opened "Code-named DHS-era UAP program revealed by AARO." AARO's memo "History and Origin of KONA BLUE" (released April 2024) says it was a prospective Special Access Program proposed to DHS leadership that "was never approved or formally established" and "never received any materials or funding". The entry now says so.
The Papoose Lake wiki entry stated, under the heading "Smoking-gun date", that on "May 23, 1989, exactly 8 days after Bob Lazar’s first anonymous TV appearance, the US Department of the Interior modified the geological map of Papoose Lake", removing the road to the east side of the lake. That account comes from Jesse Michels's American Alchemy episode with Lazar (2026); the entry now attributes the date, the eight days and the road to that episode, under the heading "Key date". The Bob Lazar entry listed the same claim and was changed the same way.
The Westall Two-Channel Bench said of the ring of flattened grass that the VFSRS's "photographs and notes can no longer be found" and that "the trace evidence is a memory of a photograph". One VFSRS photograph was printed with Paul Norman's "On the UFO trail" in the Australian Bulletin of December 2000, Charlie Wiser's Westall series reproduces two, and Ross Coulthart's In Plain Sight prints one captioned as Joy Clarke, aged 12, standing in the circle. The bench now says so.
25 September 2026
The wiki entry on Pat Price's Sugar Grove session said Price was given only the facility's coordinates, drew it accurately and read code words from inside a locked safe. The 1977 account of the CIA project officer, Kenneth Kress, in Studies in Intelligence, says the coordinates were those of "a vacation property", that the installation was found "a few miles from" it, and that both subjects described "a military-like facility"; he calls the evaluation "as usual, mixed" and mentions no safe. H. E. Puthoff's 1996 paper names the site. That entry and the Sugar Grove entry now follow these accounts. The Patty Hearst entry said Price picked the suspect out of a Berkeley police mug book, "Confirmed." The CIA memorandum of 27 November 1974 says his help was sought by "the Alameda police" and that he identified "people later proven to be involved", and names no mug book; the entry now reports the memorandum.
The Dyson bench cited Meng et al. as "(2018), ApJ 853, 1" and, with Boyajian et al. 2016, credited that paper for the dips and their "chromatic, dust-like extinction". The paper is Meng et al. 2017, ApJ 847, 131, and it measured the star's long-term fading and the missing infrared excess; the colour of the dips themselves is Boyajian et al. 2018, ApJL 853, L8. The claims table and the sources now cite both.
The Crust Displacement and Piri Reis benches credited the continuous 800,000-year EPICA Dome C ice record to the EPICA community's 2004 paper in Nature, "Eight glacial cycles from an Antarctic ice core". That paper reaches about 740,000 years; the 800,000-year record is Jouzel et al. 2007, Science 317. Both source lists now cite it.
The Nuclear Lure bench cited the 2023 Scientific Reports study of UAP report geography as "Medina & Brewer"; it has a third author, S. M. Kirkpatrick. The bench also called the Pantex document "the Pantex incident report of 2026" and wrote "2026: Pantex files an institutional incident report". The report describes a September 2015 incident at the plant and was released under PURSUE in 2026. The bench now says so and cites the full report, DOE-UAP-D005.
The Sinai Fallout bench cited the Greenland ice-core chronology it relies on as Rasmussen et al. 2006, "A new Greenland ice core chronology for the Holocene", with a maximum counting error of "~1 % of age", and dated the 2024 BC layer to ±40 years. The Rasmussen paper's title ends "for the last glacial termination"; the Holocene part of the chronology is Vinther et al. 2006, JGR 111, D13102, and its published data file gives a maximum counting error of 6 years at the 2024 BC layer. The bench now says ±6 years.