A documented history from the transcript archive

The archaeologist and the cards

Adam McBride features on twenty-four programmes in this archive between April 2021 and February 2025, and the shows billed him three different ways over those four years: first as a podcaster, then as an NFT archaeologist, then as a man from Emblem Vault. He is the first guest in World Crypto Network's own Curio Cards series — which makes this archive both a record of his work and a participant in the story it describes. This report separates what he established from what he claimed and what other people concluded, and it keeps the host's recollections apart from the documented record, because the host co-created the project at the centre of it.

Built from the WCN group — World Crypto Network, Mad Bitcoins, Today in Bitcoin, The Bitcoin Group, The Bitcoin Talk Show, Ugly Old Goat: 6,497 local transcript records, 30,351,458 words, 4,206 distinct broadcasts — and 194 matching tweets in the host's own Twitter/X archive · 24 programmes feature or discuss him; 115 broadcasts mention Curio Cards · Every priority claim is scoped to that corpus (scope statement) · Outside-archive findings researched and linked separately · Compiled 2026-10-19, rescoped to the WCN group 2026-10-21 · Rev. 2

01The method, and four ways this nearly came out wrong

Everything that follows rests on corpus.py against brain.db, and the counts were hand-read rather than lifted from a total. Four separate traps sat in this particular query, each of which would have moved a headline figure without raising an error.

The name is clean, and the token is not

Unusually for this corpus, whisper-base gets the name right. The alias machinery was probed before anything was trusted, as the method requires, and the mangles are simply not there:

That last pair is the useful result. The reversed order returns zero, which establishes that the tool phrase-matched rather than ANDing two common tokens — so adam mcbride returning exactly the same 41 rows as mcbride is a real finding: every occurrence of the surname in this corpus is immediately preceded by “Adam”. There is no other McBride on these programmes. No alias group was added to _scripts/corpus/aliases.json local only — not deployed, so no link, because the bar there is observed variants and there are none.

The token is still not clean. Searching all sources rather than broadcasts, the string returns eighteen further entries in the host's own media library, none of them this man: Christian McBride the bassist, Martina McBride, a Chick Corea record, a Joshua Redman record, an Ornette Coleman record. This is auger-the-drill-bit from CORPUS.md §5 wearing a surname. It does not corrupt the broadcast counts here because those were read per broadcast, but any session that quotes a bare total for this name will be wrong by eighteen.

The episode a name search cannot see

This is the find that matters methodologically, and it is the Junseth problem from PRIORITY-METHOD §5 with the opposite cause. There, the canonical spelling of a man's name failed to return his first mention because the transcriber mangled it. Here the spelling is perfect and the surname is simply never spoken.

Curio Cards #LIVE at Christie's Auction House, 1 October 2021, is 80,916 characters of live commentary on the auction of a full Curio Cards set. It is the most substantial single Curio Cards programme in the archive. The host addresses a participant called Adam seven times. “McBride” occurs zero times. The episode's own machine-written entity list records the participants as “Adam, Daniel, Robeck, Thomas…” — first names only, because that is all the audio contains.

It was found by a different route: it is a member of the Adventures in NFTs playlist, sitting four places from his interview, which is CONVENTIONS §4c's rule earning its keep — the owner's own grouping finding a programme that no title or name filter reaches.

And the attribution consequence is strict. CONVENTIONS §6 allows tier A where a hand-off is answered, and there is one: “So Adam, how's it going down there in the Discord?” followed by a reply. That establishes the speaker is the person called Adam. It does not establish the surname. The grounds for thinking it is McBride are circumstantial and good — the programme is about Curio Cards, the playlist places it beside his interview, he is the Curio Discord member the host interviews about Curio Cards in the adjacent episodes, and the other named participants are the Curio artists he had just interviewed — but circumstantial grounds are tier B, and nothing is named at tier B. So every quote from that episode in §05 runs unnamed, under the flag, and the reader can see the reasoning rather than inherit a conclusion.

Two of the 24 are republications

Two programmes in the set are rebroadcasts of other programmes in the same set, and both say so on tape:

• Curio Cards 6th Anniversary Retrospective (2023-05-10) opens with a live read of OpenSea statistics and then replays the May 2021 McBride interview in full. The interview text appears verbatim in both transcripts. Its ten “McBride” mentions are the 2021 interview's, counted a second time.
• The Bitcoin Group — Tenth Anniversary Spectacular (2023-10-19) states it plainly — “as streamed a year ago on March 18th, 2022… with the Bitcoin Group number 300” — and carries #300's cold open, including his roll-call.

Treated as appearances they inflate him by two. Deleted they would hide something real: the network chose to rebroadcast him twice, which is an editorial fact about how it valued the material. Both are kept, counted once each in the 24, and marked in the appendix.

Four of the 27 search hits are Shorts, not broadcasts

corpus.py search mcbride reports 27 episodes. corpus.py first and spread, which exclude the short lane by default, see fewer — and the gap is exactly four: four WCN Shorts of 5, 6 and 10 February 2025, machine-cut from The Bitcoin Group #440 of 1 February and given topic titles (Elon Musk's Twitter Play, Equal Sentencing, Sam Bankman Fried's Parents Seek Trump Pardon!, Roger Ver and the Pardon Controversy). Each inherits a single “Adam McBride, what do you think” from the parent episode. Per CORPUS.md §2c a Short is never a broadcast, so the figure in this report is 23 named broadcasts, not 27 — plus the unnamed Christie's stream, giving 24 programmes.

02Who he is, and the three names the shows gave him

He enters the corpus on The Bitcoin Group #254, 16 April 2021, as a billed panellist. He had been in the Curio Cards Discord for a month by then, and in the host's Twitter archive for a month and a day. He is not in the corpus at all before that date.

What he was billed as is on tape in each case, in the programme's cold open, immediately followed by his own reply — which is CONVENTIONS §6 tier A: a hand-off this quote answers.

1. 2021-04-16 → 2021-12-31 · five cold opens “Adam McBride from the Adam McBride show.” The podcaster. On #254 he answers “Let's go.”; on #265, “Bro, I am not muted. Great to be here.”
2. 2022-03-18 and 2022-05-20 “Adam McBride, NFT archaeologist.” The title arrives eleven months after he does. On #300 he answers “What's up everyone?”; on #308, “Bro, I want to name like the Raptor, man.” He does not correct it either time.
3. 2023-03-17 → 2025-02-01 · three cold opens “Adam McBride from Emblem Vault.” The employer replaces the vocation. On #361 the billing is “Adam McBride, from Emblem, vault, ordinals, ordinals, ordinals”, and in that episode he reports Yuga's ordinals drop and the secondary sales “all on Ethereum using Emblem Vault”.
4. 2024-05-10 “we have Adam McBride, the NFT historian here.” The fourth label, used once, by Thomas, on the Curio Cards 7th Anniversary. It is the most accurate of the four, for the reasons §04 and §05 set out.

Outside the archive, the identity resolves consistently. He publishes at adamamcbride.medium.com, posts as @adamamcbride, and runs The Adam McBride Show (Apple ID 1410811576, launched 2018 — before NFTs, which matters: the podcast was not built for this). Emblem Vault's own post dates his arrival precisely: “When Adam McBride and Jake Gallen joined Emblem Vault on November 1st, 2022…” — which places the on-air billing change in March 2023 about four months behind the fact.

He also wrote a book, and this archive holds a record of it that is better than a mention: see §08.

03What he established: the published catalogue

This section is outside the archive. His Medium sitemap is the authoritative inventory of what he published and it is complete: seventeen posts, 4 May 2021 to 29 November 2022, of which nine surface a project and the rest are how-tos and follow-ups. The list is in §10. Nine project posts, nineteen months, then nothing.

Two absences in that list are findings. There is no MoonCats post and no Curio Cards post. His output on Curio Cards was the podcast, not an article — which is why this archive, and his own show, are where that work survives. Any claim that he “published” the Curio or MoonCat rediscovery is unsupported by his own publication record.

Pixel Map — the one he says he found

“Pixel Map — Lost 2016 NFT Project”, dated 23 August 2021 on the page. This is the only item in the catalogue where he describes himself as the finder, and he does it in passing rather than as a boast:

The post also establishes the project facts he is relied on for — deployed 17 November 2016, 3,970 tiles, 2 ETH each, “or about $20 in 2016” — and records that “@LeonidasNFT excellent NFT history infographic shows exactly ZERO projects in 2016”, which is the claim to significance. Independent corroboration exists and credits him by name: Crypto Briefing, three days later, reports that “The contract resurfaced thanks to a Medium post from NFT enthusiast Adam McBride”.

It is also the post that documents his conflict of interest, in his own words, including an apology. That is §07.

Terra Nullius — the biggest claim, and he gave the credit away

“Terra Nullius — The First NFT on Ethereum”, dated 27 September 2021 on the page. It is the single most consequential claim associated with his name, and the first substantive sentence hands the find to somebody else and hedges it:

The post carries a retroactive correction banner at the very top, which is itself part of the record:

Four years on, the claim has hardened in his favour without the hedge: Guinness World Records now records “First Ethereum NFT” as Terra Nullius, 7 August 2015, by the Reddit user Semiel, 4,000 claimable blocks. The original 2015 Reddit announcement is still at its permalink, with Wayback captures running from 2018 to 2024. Its state right now could not be confirmed for this revision: Reddit answers every automated request with an anti-bot challenge, and old.reddit.com with a logged-out login redirect. Neither is evidence of removal and neither is evidence of survival, so this report does not assert that it is live today — only that the URL is the real post and was captured repeatedly over six years. What seems to be has become is, and nothing in the record did that work; attention did. The post's own disputed sentence is §06.

The rest of the catalogue

Full list with links in §10. Of the nine project posts, the ones that recur in secondary coverage are EtherWaifu (10 February 2022 — and the one he describes on this archive as a find he made himself, with a friend), CryptoFighters (“the first play-to-earn NFT”), Aether (a 2018 metaverse, and the one This American Life picked up), and the 28 September 2021 post on an unnamed “first female art project on Ethereum”. His Medium appears dormant after November 2022.

And the pattern of §04 shows up in the catalogue itself, not just in the famous cases. Reading his profile index, the credit is routinely somebody else's in the standfirst. On MyBlockchainCorner: “NFT Archeologist Brain Genius found this fantastic 2017 NFT project and it absolutely blew up.” On Terra Nullius, Wilt Chamberlain. On the 2017 art project he hedges his own claim in the subtitle — “what I believe to be the first female art project on Ethereum”. Where he does claim the dig, the phrasing is impersonal: “NFT archeological digging has turned up another long-lost project” (Aether), “NFT archaeology has uncovered the very first in-game NFTs on Ethereum” (Etherization). Pixel Map's “Since I first rediscovered” is the exception in his whole catalogue, which is why §03 rests the finder claim on it and on nothing else.

04What he did not find — in his own words, on this archive

This is the section the brief most needed, and the archive answers it better than the outside sources do, because he is on tape here describing his own arrivals before the story had settled into a shape.

On Adventures in NFTs #1, 25 May 2021 — a single-guest interview, so every quote below is tier A under CONVENTIONS §6 — he walks through three separate rediscoveries, and the mechanism is the same each time.

He got five of them, and explains why only five: “If you don't put in the right gas fee, you're getting front run… I'm just hitting go. I'm hitting buy. So all of mine are average grass gas.” (“grass gas” is the transcriber for gas.) The account he gives of the day is domestic rather than archaeological — he and his son, homeschooling, five tabs open, a shared Google Doc, “get the yellow one. Get the yellow one. I think it's rare.”

On Pixel Map, in the same hour:

And on Curio Cards:

Two weeks earlier, asked point blank to introduce himself at the Curio Cards 4th Anniversary Party, he had been even plainer. The host's question is the hand-off: “Adam McBride, who are you? Say hello.”

He begins to claim a status — “one of the first people to kind of” — breaks off mid-sentence, and replaces it with the DM. The second half of that quote is the work he actually did, and it is §05.

MoonCats: four names, shared credit, from the project itself

MoonCatRescue's own About page is the primary document and it was read directly for this report. It credits four people, and it is the source that ties his pseudonym to his account:

On that page the name Adam McNFT is hyperlinked to x.com/adamamcbride. That link is the whole basis for identifying the two, and this report could not establish whether “Adam McNFT” was a former handle of his or a nickname the MoonCat team applied.

Two positions, and they do not reconcile. The project credits him as one of four. SuperRare's own profile of him credits @ETHoard and quotes McBride as someone the event happened to: “And then the MoonCats happened in March… man, and I was in the rabbit hole 10, 12 hours a day hunting, because I wanted to find my own MoonCats.” The archive quote above is consistent with the second reading, not the first. This report does not resolve it. The defensible statement is that the project names him among four, that an independent profile of him credits another person, and that his own account on this archive describes receiving a DM.

One date is now firm, and it is useful downstream: MoonCats, 12 March 2021, from the project's own documentation.

Curio Cards: the co-creator credits Harry BTC

Here the archive holds something no outside source does — a co-creator of the project describing the rediscovery on air, three years later, with McBride present in the room.

Note what the sentence does: it names Harry BTC as the finder without qualification, and hedges McBride's role — “i think adam mcbride involved as well as others”. The same host's tweet of 12 March 2024 drops the hedge in the other direction: “Shoutout to @HarryBTC for his 2021 rediscovery of @CurioCards. Here are some screenshots I took of this historic event. (3/14/2021)” — verified live.

And McBride himself has consistently routed the credit the same way. On 2 August 2021 he published a podcast episode asking the question outright, and the host's archive holds the retweet: “Who is responsible for the Curio Cards rediscovery?🧐 According to @HarryBTC it's a shadowy figure named ‘Bob’.” Thomas replied “It was Bob.” with a link to the Church of the SubGenius. Eleven days later: “Good to see @HarryBTC getting the recognition he deserves!”

Three dates are in play and no source reconciles them.

The first dated trace in the host's own archive falls on 15 March 2021 — consistent with any of them, and the subject of §08.

05Curio Cards: what the archive actually holds

This is where his own project and McBride's research meet, and it is the largest body of material in the report. 115 broadcasts across five of the six WCN-group shows mention Curio Cards, 9 June 2017 to 2 October 2026. The shape of that is the finding:

Two mentions in the launch year, one in 2019, and then forty-two in 2021. The archive records the project's death and resurrection as a mention count without anyone having intended it to.

This is the one count the WCN-group scope moved, and it is recorded rather than quietly restated. Searched across all ten searchable shows the figure was 117 across six, with 2017 at four. Two of those 2017 mentions are on the San Francisco Bitcoin Meetup — Thomas plugging his own project from the floor of his own meetup, on 28 June and 30 August 2017 — and SFBM is outside this report's scope, so they come out: 115 across five, 2017 at two. Worth knowing that both are physically uploads on the WCN channel, classified as SFBM through shows.json nested_ids; the line this report draws is the register's, not the channel's. The 2017 row is the only cell in the table above that changed; every other year, and the first-mention date itself, is identical under both scopes.

The first mention is a month after launch, on The Bitcoin Group #145, 9 June 2017 — “So thank you for supporting artists with Curio cards. And we had 587 viewers right now.” — and it is a plug, not coverage. Thomas's own interview with Robek's World follows on 21 July 2017. Per the scope statement, “first” here means first in the WCN group, which is the only corpus this report searched. It is not a claim that no earlier broadcast discussion exists anywhere. The runner-up under the wider scope was the SFBM evening of 28 June 2017, nineteen days later, and it is out of scope rather than absent.

What he actually did: the oral history, made on this archive

The Adventures in NFTs — The History of Curio Cards playlist holds 14 entries, 13 distinct (the roundtable is listed twice), captured from the channel on 15 October 2026 with zero unavailable members. McBride's interview is episode #1. The series then works through the people he had spent those three or four days contacting: Travis Uhrig (co-founder), Daniel Friedman, Max Infeld, Robek, Yulia Babanova, Luis Buenaventura.

And the provenance runs in a loop. In episode #6, Robek describes having been on McBride's show: “so after you got on the discord and then I saw your interview on Adam McBride show.” In the 2023 retrospective, Robek says the same of another artist: “I've listened to Daniel's podcast. He did with Adam McBride on the Adam McBride show. And it's very cool.” Max Infeld, in the 4th Anniversary Party, remembers the contact itself: “When the first messch as it got was from Adam McBride and he thought I was a project coordinator. I'm like, oh no, you need to talk to these other guys.”

And on 15 July 2021, ahead of his own Twitter Space, McBride pointed his audience back at this archive: “In advance of the Curio Cards Twitter Space I wanted to share this interview I did with @MadBitcoins which I think gives…” (truncated in the host's archive). That circularity is real and it is the reason §01 insists on labelling which claims rest on this material.

The supply confusion, and a number he got wrong in a useful way

Both men are on tape in the same programme reconstructing the same detail from opposite sides. Thomas, on his own project:

McBride, immediately after, on the same moment:

The vending machine advertised a six-figure supply; the contract burned the remainder at the end of the week. Curio Cards' own documentation gives the settled figures: 30 cards, seven artists, supplies 111 to 2,154, 29,700 total, “with an estimated 4,000 currently lost or destroyed”. So the thing that made him pass was a number that had already stopped being true — and in Quartz, five months later, he put the cost of that decision at about $250,000 and called it an “epic and awesome” mistake. It is the clearest illustration in the record that being first to the contract and being right about it are different achievements.

Christie's, 1 October 2021: the hammer and the invoice

The archive holds the auction as live audio, captured through a restream of the sale, and the framing of the lot is in it:

The bidding is in the transcript and it stops at 320: “I'm selling now 320 Ethereum. Sold with you”. The room then does the arithmetic out loud — “it looks like the final bit, the final USD was one million, 35,000 dollars” — and this is where the archive and the published record part company.

One further thing in that transcript should be marked rather than used. In the minutes after the hammer, participants speculate about the buyer — “North Carolina took it. It's Mister Beast.” That is a guess made live on air and it appears to be wrong: Castor and Gerard identify the buyer as Taylor Gerring, an early Ethereum contributor, citing his own tweet. The archive's value here is that it preserves what the room believed at the time, not that it settles who bought the set.

Five birthdays, and the line that recurs in all of them

The project has had an annual programme on this network since 2021 — 4th (2021), 5th (2022, on Twitter rather than in the video archive), 6th (2023), 7th (2024), 8th (2025). The theme that recurs across all of them is not provenance; it is improbability, and it is McBride's framing. He put it to the founders in his first week:

Three years later Thomas was still attributing the framing to him by name, in his own words, on 2 May 2024: “What I really like about this video is how well it explains what @adamamcbride brought up when he first interviewed me: A startup never dies and comes back to life, but @MyCurioCards did.” (verified live.)

At the end of the 4th Anniversary Party the host gives him a title that sits precisely between the two readings in this report — “Adam McBride, you're an early discoverer of CurioCards. What are your closing thoughts?” — and his answer is about the people, not the cards: “it's far more than the cards. It's the community around the cards that is special.”

06Where the claims were contested, and where he revised them

Early-NFT provenance claims are contested as a matter of course, and his are no exception. What the record shows is a pattern in how he handled it.

The CryptoPunks interval: four numbers, and his own correction is the tightest

The correction, which the host retweeted and which therefore survives in his archive: “Update: Curio Cards launched on May 9, 2017, more than a month before CryptoPunks, which launched on June 23, 2017”. 9 May to 23 June is 45 days, so his own figures land on Castor and Gerard's six weeks and not on his own earlier “two weeks”. The launch date he gives for Curio Cards matches the project documentation exactly.

The ordering, revised twice in five months

1. 9 May 2021On the day of the 4th anniversary, he posts a ranking the host retweets: “If y'all don't know- 1st - @etheria_feed 2nd - @MyCurioCards 3rd - @larvalabs CryptoPunks. Now that said, Punks clearly…” (truncated in the archive). Etheria first.
2. 25 May 2021On this archive he narrates the same ordering as a sequence of displacements, and is explicit that it is unsettled: “So this is now the first art NFT project. As far as we know, there's still a possibility that somebody else could be, you know, sneak in there.”
3. 27 Sep 2021His own Terra Nullius post displaces Etheria — and credits the find to Wilt Chamberlain, with the hedge “what seems to be”. His own prediction that nobody would sneak in was wrong within four months, and the person who proved it wrong was his friend.

The competing positions, attributed: @LeonidasNFT's timeline, as published by One37pm, draws the distinction that keeps both alive — Terra Nullius “the first NFT on the Ethereum blockchain with a mint date of August 7, 2015”, Etheria v1.1 “the first NFT on Ethereum with a transfer function”. That Terra Nullius therefore lacks a transfer function is this report's inference from the pair, not something the source says; it states the one as a distinction and does not state the other as an absence. Castor and Gerard take the art qualifier seriously: Etheria predates Curio Cards but sold plots of virtual land, “not art”. Quartz notes that the oldest candidate of all predates Ethereum entirely. The claim “first” is doing different work in each of these and the sources are not in disagreement so much as answering different questions — which is itself the most useful thing to say about early-NFT provenance.

Card 3, then Card 1, within 48 hours

The sharpest documented self-correction in the record, and he flagged it himself. Both texts survive as retweets in the host's archive and both are truncated there; the ellipses below are the archive's.

• 18 July 2021. “Curio Cards Facts #4🤔 Cards 1, 2, 3 were issued in that order. Making Card 1 the first Art NFT on Ethereum. BUT Card 3…”
• 19 July 2021. “🚨UPDATE - One of our Sherlock Discord members investigated further and found that Card 1 (Apple) actually WAS the first c…”

A day earlier he had also published the detail the correction turns on: “Curio Cards Facts 🤓 Cards 1, 2, and 3 - the first set - were released in that order with Card 1 being four minutes befor…” Because all three are truncated, this report can establish that he made a claim about card ordering, published a correction to it within 48 hours, and credited the correction to someone else — and cannot establish from this archive what the intermediate claim about Card 3 was. The un-truncated texts are on his own timeline and were not read for this report.

The 17b misprint: he is the named source, and he disclaims certainty

Castor and Gerard's account of the misprint rests on him, and the quote they take is a disclaimer:

The archive corroborates the card's existence and the set-size question rather than the cause. At the top of the 6th Anniversary Retrospective: “There are 31 unique Curio cards that's the 30 original cards, including the 17 B misprint. It was a contract that went bad.” The project's own documentation numbers the cards 1 to 30 and treats 17b separately, with its own wrapper endpoint. So “30” and “31” are the same set counted two ways, and the Christie's lot was described in the room as “the full set of 30 cards and the miss print 17 B” — which is the honest formulation.

Two things this report could not establish: whether cards 18 and 19 were also initially misprinted, which appears in search summaries attributed to Uhrig but could not be sourced to a page that was read; and McBride's claim, reported secondhand by Castor and Gerard, that Curio Cards was the first project to use IPFS. The project's own documentation lists IPFS-hash embedding among the design elements later standards share, which is a weaker and different claim.

07The conflict of interest, disclosed and otherwise

Nothing in the record alleges impropriety and this report alleges none. What the record does show is a man whose research and whose trading were the same activity, who said so, and whose framing of that shifted as the money got larger.

His first recorded word for the practice was not archaeology. On 15 March 2021, three days after MoonCats and within a day or two of Curio, the host retweeted him: “What I learned today investigating - a.k.a FOMOing - Curio Cards. The project was done by @travisformayor who was active…” The joke is the disclosure. Archaeologist arrives on this archive's microphone a year later.

On Pixel Map he disclosed, apologised, and carried on buying, all in the same post.

Ten minutes is the whole disclosure regime, stated plainly, and he volunteered it. The same post warns readers off three dead contract addresses in capitals. On Terra Nullius a month later he discloses the same way — “Wilt and I minted another 2 NFTs” — in the post that publicises the contract.

He also became a dealer, and said so. 1 August 2021: “Historic day for Curio Cards!!👀 I am honored to have assisted in the first sale of a complete set. Cards 1-30 plus a 17B…” 10 November 2022: “We have a new full set @MyCurioCards holder🚀🎉 I assisted the collector in acquiring his final card…” In Quartz he put the first of those at 100 ETH. And on 11 September 2021 he described his own research Discord in market terms: “The greatest NFT Archeologist Discord in the world. Only one Channel - Alpha”.

The framing moved. SuperRare's profile, whose body text is plainly from around August 2021, reports that “he's yet to make money from any of his archeological digs, save some gifted NFTs from those he's helped”. By Quartz in October 2021 he has bought and sold Curio Cards and facilitated a six-figure sale; by Morning Brew in March 2022 it is reported as his full-time job. Those are not contradictions — they are eight months apart in a market that moved enormously. The point is that the “I don't profit from this” framing is specific to mid-2021 and should not be carried forward.

The structural critique belongs to Castor and Gerard and is attributed to them:

Set against it, from the same practice: a developer who had made about $300 from a 2016 side project was reachable, relaunched his platform, and — per SuperRare — earned over $3.4 million; an artist in Indonesia got what McBride, on this archive, called “a life changer”. Both readings are supported by the record and neither cancels the other.

08Thomas's own stake — recollection and record, kept apart

Thomas Hunt co-created Curio Cards, which is the reason this report exists and also the reason it has to label its own evidence. Wikipedia and Quartz both give the founders as Travis Uhrig, Thomas Hunt and Rhett Creighton, and he confirms it himself on air, introducing Uhrig in the 2023 retrospective: “about four years ago, you and I and Rhett Creighton, work together on a little project called Kiro cards” (“Kiro” is the transcriber).

Recollections are labelled as recollections. Everything in this report where Thomas describes the project's own history — the Harry BTC credit in §04, the 100,000 supply in §05, the vending-machine fair-launch argument — is an on-air recollection by a participant. It is good evidence about what a founder believed and said, on a date, and it is not independent corroboration of itself. Where the documentary record exists it is cited alongside and the two are not merged.

The documented record

Three items in the archive are records rather than memories, which is a different evidential class.

1. The first dated contact, 15 March 2021. The host's own quote-tweet of McBride's investigation thread. It is live, and it is the best single line in the whole file:

It quote-tweets McBride's thread of the same day, which is also live. A co-creator reading a stranger's reconstruction of his own failed startup and calling it archaeology — in the same week the word started being used about McBride — is the moment the two halves of this report meet, and neither man was performing for an audience when it happened.

The following day Thomas was supplying primary material into the reconstruction: “For @adamamcbride, I don't think it was ever released, but here are some screenshots from the rough draft of the legendary #CurioCards' Plaidpaper” — and correcting the record on card scarcity with a link to a community tool. The research was collaborative from the second day.

2. The Goodreads record. The host's media library holds McBride's book as a read-and-rated entry, and this is a dated record, not a recollection:

18 April 2022 is two weeks after the two of them were on the historical-NFT panels at NFT Miami, which the host's tweets of 3 April 2022 document. This is the same class of find as the media-library record behind the Richard Heart report's §06: an independently-kept log that dates an engagement nobody set out to document. It also settles a book whose own site could not be read for this report.

3. The 194 tweets, and what they are good for. 99 from 2021, 60 from 2022, 17 from 2023, 17 from 2024, 1 from 2025 — a curve that tracks the Curio Cards mention curve in §05 and then falls off a cliff. The dense March–August 2021 run is the best available record of the rediscovery as it happened, because it is contemporaneous and timestamped. Its limits are real and stated in §10: most of the McBride material in it is retweets, which are truncated in the export and not linkable by permalink.

And one item is dead, and is kept. On 10 May 2021 Thomas replied to the thread where McBride, the artists and the founders were establishing the record, with the line “/casually puts the date on there /documents history”. Checked for this report and ✕ deleted — no longer resolves: https://x.com/MadBitcoins/status/1391560272788168704. A tweet about documenting history, which is no longer there, in a report about reconstructing a project from what survived. It is kept per CONVENTIONS §3 because the disappearance is part of the record.

09What could not be established

Named rather than buried.

• Whether the “Adam” on the Christie's livestream is Adam McBride. The grounds are strong and circumstantial; the surname is not on tape; tier B names nobody. The quotes from that episode run unnamed. Resolving it needs the video's description, participant list or a frame — none of which this compilation held.
• Who rediscovered Curio Cards, and on what day. Four candidate accounts (@HarryBTC, @DieAping, “Bob”, and the unnamed Twitter contact McBride describes) and three candidate dates (13 or 14 March 2021, with MoonCats firm on the 12th). No source reconciles them, and the two that are most load-bearing — Thomas's and McBride's — disagree with each other by a day.
• Whether “Adam McNFT” was his handle or a nickname. Only mooncatrescue.com's hyperlink ties it to his account.
• His current title at Emblem Vault. Reported as CSO on his own X bio; no company page, release or filing naming him was found. Treated as self-reported.
• The un-truncated text of his retweeted claims. The host's export truncates retweets, so several of the claims in §06 can be established as having been made and corrected but not quoted in full. The originals are on his own timeline.
• Whether Curio Cards was the first project to use IPFS. A McBride claim reported secondhand; the project's own documentation makes a narrower one.
• Whether cards 18 and 19 were also initially misprinted. Appears in search summaries attributed to Uhrig; could not be sourced to a page that was read.
• The provenance of the Terra Nullius wrapper contract. His post asserts “Clearly, the team behind the project recently built the wrapper”; at least one later account describes it as an anonymous third party's, deployed the day before his post. That source could not be read (client-rendered, empty body) and is not cited as establishing anything.
• The This American Life segment. Thomas's tweet of 4 May 2022 congratulates McBride and Ken Erwin on appearing; episode 769, The Reluctant Explorer, is the candidate. The transcript was not read and the segment was not confirmed.
• Anything outside the WCN group holds. Twelve of the archive's eighteen registered shows were not searched — see the scope statement. This is a boundary, not an outstanding risk: a claim scoped to six named programmes cannot be falsified by a show it never claimed to cover. It can be extended by one, and the four searchable shows outside the scope were checked for McBride before they were dropped and hold nothing about him.
• Whether he is in the archive before April 2021 under a form not probed. Six spellings returned zero and the phrase test establishes the surname is never used without “Adam”; a first-name-only appearance before 2021, like the Christie's one, would be invisible to every method used here.

10Sources

Two kinds of material went into this report and they are kept apart here, because the separation is the report's spine. Archive is the local transcript corpus, the host's own Twitter export and his media-library records — §01, §02's cold opens, §04's on-air quotes, §05 and §08. Outside material is his own publications, project documentation and press — §02's identity facts, §03, §06 and §07. Nothing from the second pile supports a claim about what was said on air, and nothing from the first supports a claim about what he published.