Bitcoin was already using outside information to settle valuable agreements before most people had heard of the problem. But much of that history is now missing, and the surviving record can point researchers in the wrong direction. Before Ethereum made the “oracle problem” a household term, Bitcoin already had oracles serving as feeds, key-release services, federated signers, and arbiters that carried real value on the main chain.

Before Ethereum made the oracle problem a household term, Bitcoin already had oracles serving as feeds, key-release services, federated signers, and arbiters carrying real value on the main chain. This study traces their use and the changing evidence of oracle activity from early days through July 2026.

The study combines betting records, the Bitcoin blockchain, documented service keys, archived websites, and later public announcements and attestations. Two findings stand out. Activity depended not only on what happened, but also on the role of public keys, the information disclosed by transactions, and the external records available.

The first recorded bet was placed on 2 February 2014. It was a one-unit wager on whether Seattle won the Super Bowl. The activity was concentrated in 2014, with only 44 bets recorded after that year, and the last recorded bet appearing on 12 June 2017. The activity arrives in sharp bursts rather than a steady stream: one towering surge, a smaller cluster later in twenty fourteen, and only isolated traces afterward.

Activity was also concentrated among a small number of services. The largest individual feed accounted for 22.5 percent of matched value. Two services together accounted for 98.2 percent of matched value, showing that the visible market was dominated by a very small part of the system.

The later records came from archived announcement and event pages captured between 2019 and 2023. They produced 911 distinct announcement records. Of those records, 73.9 percent were associated with attestation records in the recovered pages, and recorded event dates were concentrated in 2021 and 2022.

But the historical trail is incomplete. Early contracts remain on-chain, while many descriptions of the events they concerned have disappeared, making the surviving record harder to interpret. For modern contracts, public announcements can survive even when the contracts using them cannot be identified on-chain.

Public registries can also give a misleading picture of use. In Counterparty, 95 percent of pre-2018 sources that declared an oracle fee were never bet on. So a published service may show that information was offered, without showing that anyone actually used it in a contract.

The findings show how contract design shapes the evidence available for measuring use. Revealed scripts and documented keys can connect transactions to particular services, while spending paths that disclose no script affect attribution.

Announcements and attestations provide a complementary record of the information published by oracles, beyond the bets and settlements visible in older systems. Interpreting the evidence requires distinguishing publication activity from the contracts that consume it, and considering how that relationship changes across implementations.

For people trying to understand Bitcoin’s economic history, attribution depends on what scripts and keys are revealed, because shared structures and undisclosed spending paths can obscure which service was involved. DLC announcements and attestations add a complementary record of what oracles published, but researchers must distinguish that publication from contracts that actually consume it.

Bitcoin’s past oracle use was real, but the chain alone cannot show the whole story. What survives depends on how contracts were built, what transactions reveal, and whether the surrounding records were preserved.