Inscription (Bitcoin)

What Is a Bitcoin Inscription?

A Bitcoin inscription is a piece of arbitrary data: an image, a text file, a JSON object, or even an HTML page: permanently embedded inside a Bitcoin transaction. Unlike traditional NFTs on other blockchains that often store metadata on external servers, inscriptions live entirely on-chain within Bitcoin's witness data field.

Inscriptions were introduced by Casey Rodarmor as part of the Ordinals protocol, which launched on Bitcoin mainnet on January 21, 2023. The protocol assigns a unique serial number to every individual satoshi ever mined, making each one individually trackable. An inscription attaches data to a specific satoshi, turning it into a unique digital artifact that travels with that satoshi across transactions.

The very first inscription (Inscription #0) appeared on December 14, 2022, during a 36-day mainnet testing period before the protocol's official launch. By early 2024, the total count had surpassed 60 million inscriptions, and by September 2024 it exceeded 70 million.

How It Works

Inscriptions rely on two Bitcoin upgrades: SegWit (activated August 2017) and Taproot (activated November 2021). SegWit separated signature data from the main transaction body and gave witness data a 75% fee discount. Taproot added Tapscript, which removed the previous 80-byte limit on data that could sit inside a script. Together, these upgrades made it practical to store up to approximately 4 MB of data in a single block's witness space.

The Envelope Format

Inscription data is wrapped in what the protocol calls an "envelope": a sequence of opcodes that stores data without affecting script execution. The envelope uses OP_FALSE OP_IF to create an unreachable code path. Because the condition is always false, Bitcoin nodes never execute the data pushes inside it, but the data is still permanently recorded in the witness field.

OP_FALSE
OP_IF
  OP_PUSH "ord"           # Protocol identifier
  OP_PUSH 1               # Content-type tag
  OP_PUSH "image/png"     # MIME type of the inscription
  OP_PUSH 0               # Body tag (marks start of content)
  OP_PUSH <raw image data> # Actual inscription content
OP_ENDIF

Because envelopes are effectively no-ops, they do not change the semantics of the script they appear in. This means an inscription can be combined with any other locking script without affecting how that script functions.

Commit-Reveal Process

Creating an inscription is a two-step process involving two separate transactions:

1. Commit transaction: a Taproot output is created that cryptographically commits to a script containing the inscription content. The data itself is not yet visible on-chain at this stage.
2. Reveal transaction: the commit output is spent using a script-path spend, which reveals the full inscription data in the transaction's witness field. The inscription is now permanently recorded on the blockchain.

This two-phase approach prevents front-running: nobody can see the inscription content and attempt to inscribe it first, because the content only appears when the reveal transaction is mined.

Why Witness Data?

The SegWit upgrade introduced a concept called weight units that treats witness data differently from non-witness data. One byte of witness data counts as 1 weight unit, while one byte of non-witness data counts as 4 weight units. This effectively gives witness data a 75% discount on fees, making it the most cost-efficient place to store large amounts of data in a Bitcoin transaction.

Types of Inscriptions

The inscription ecosystem has expanded well beyond simple images. Several distinct categories have emerged, each serving different purposes.

Image and Media Inscriptions

The original and most recognizable type: images, audio, video, and other media files embedded directly on-chain. These function similarly to NFTs on other blockchains, but with the key distinction that the entire file lives on Bitcoin rather than on an external server. Common formats include PNG, JPEG, SVG, and WebP. The maximum size for a single inscription is constrained by Bitcoin's block weight limit, practically capping individual inscriptions at under 400 KB in most cases.

BRC-20 Tokens

BRC-20 is a fungible token standard created in March 2023 by a pseudonymous developer called Domo. BRC-20 tokens use JSON-formatted inscriptions to deploy, mint, and transfer tokens. Each operation is a separate inscription containing a small JSON object:

{
  "p": "brc-20",
  "op": "mint",
  "tick": "ordi",
  "amt": "1000"
}

BRC-20 transfers require two on-chain transactions: one to inscribe a transfer instruction, and another to actually send it. This two-step process generated significant transaction volume and contributed to network congestion during peak activity in 2023 and 2024.

Recursive Inscriptions

Recursive inscriptions reference data from previously inscribed content instead of embedding everything in a single transaction. An inscription can include a reference like /content/<inscription_id> to pull in data from another inscription, enabling composability. This approach dramatically reduces costs for complex content: a generative art collection, for example, can inscribe a shared JavaScript library once and have hundreds of individual pieces reference it, rather than duplicating the code in every inscription.

Runes

Runes is a fungible token protocol also created by Casey Rodarmor, launched on April 20, 2024, at the exact block of Bitcoin's fourth halving. Unlike BRC-20, Runes does not use inscriptions at all. Instead, it encodes token operations inside OP_RETURN outputs and works natively with Bitcoin's UTXO model. Rune transfers complete in a single transaction, addressing the efficiency issues of BRC-20's two-step process.

Fee Impact and Block Space Economics

Inscriptions introduced a new category of demand for Bitcoin block space. By the end of 2023, users had paid over 5,300 BTC in total inscription-related fees. By September 2024, that figure had grown to more than 6,900 BTC.

During peak inscription activity, transaction fees spiked significantly. The fee market saw periods where inscription demand pushed average fees above $10 per transaction, pricing out smaller financial transfers. During the most intense periods in 2025, inscription-related fees in some blocks actually exceeded the 3.125 BTC block subsidy itself.

This dynamic has implications for Bitcoin's long-term security budget. As block subsidies continue to decrease through successive halvings, transaction fees must eventually sustain the network's security. Inscription demand has demonstrated that non-financial use cases can contribute meaningful fee revenue: by 2026, transaction fees have stabilized around 15% of total miner revenue, up from single-digit percentages in prior years.

The Block Space Debate

Inscriptions have sparked one of Bitcoin's most heated debates since the block size wars.

Arguments Against Inscriptions

• Inscriptions compete with financial transactions for limited block space, potentially increasing fees for users who need to send small-value payments
• Storing images and media on a financial settlement layer adds permanent data that every full node must download and store, increasing the cost of running a node
• The witness discount was designed to incentivize SegWit adoption, not to subsidize data storage: inscriptions exploit this discount for an unintended purpose
• BRC-20 tokens in particular generate high transaction volumes with relatively low economic value, contributing to mempool congestion

Arguments For Inscriptions

• Bitcoin is a permissionless network: if a user pays the required fee, their transaction is valid regardless of its content or purpose
• Inscription fees provide additional revenue for miners, strengthening Bitcoin's long-term security budget as block subsidies decline
• The data uses witness space that is already discounted: removing inscriptions would not meaningfully increase capacity for financial transactions
• Inscriptions bring new users and developers into the Bitcoin ecosystem, expanding its cultural and technical reach

Why It Matters

Inscriptions represent a fundamental expansion of what Bitcoin can be used for beyond simple value transfer. Whether you view them as innovative or wasteful, they have already had lasting effects on the ecosystem: driving new token standards, reshaping the fee estimation landscape, and forcing the community to reckon with how a permissionless network handles competing uses of scarce block space.

For Layer 2 solutions like Spark and the Lightning Network, inscription-driven fee spikes on the base layer reinforce the value proposition of off-chain payment networks. When on-chain fees rise due to inscription activity, moving everyday payments to Layer 2 becomes more economically compelling.

Risks and Considerations

Blockchain Growth

Every inscription adds permanent data to the Bitcoin blockchain. While individual inscriptions may be small, millions of them accumulate. This increases the storage requirements for full nodes and extends the time needed for initial block download. Node operators with limited storage may be forced to use pruning, which reduces the number of archival nodes on the network.

Content Permanence

Data inscribed on Bitcoin cannot be removed or modified. This permanence is a feature for legitimate use cases, but it also means illegal or harmful content inscribed on-chain exists in perpetuity. Every full node operator stores and serves this data, raising legal and ethical questions in various jurisdictions.

Fee Volatility

Sudden surges in inscription activity can cause dramatic fee spikes, catching regular users off guard. A user broadcasting a time-sensitive financial transaction during an inscription wave may need to pay significantly higher fees than expected. Effective fee estimation becomes more challenging when block space demand is driven by non-financial activity with different urgency profiles.

Regulatory Uncertainty

The classification of inscriptions and inscription-based tokens like BRC-20 remains unclear in most jurisdictions. Depending on how regulators categorize these assets, creating, trading, or holding inscriptions could carry compliance obligations that the ecosystem is not yet equipped to handle.

This glossary entry is for informational purposes only and does not constitute financial or investment advice. Always do your own research before using any protocol or technology.