LibraryStorage and networks2018Design paperCorpus record
Basic Attention Token: Blockchain Based Digital Advertising
Basic Attention Token. Brave Software.
Brave's 2018 paper for an advertising unit that pays publishers and users from a measured attention event inside the browser, rather than from a chain of third-party trackers. The browser is load-bearing. The token is the unit of account in the paper's payment flow.
The Basic Attention Token paper proposes that a browser measure attention and pay publishers from an advertiser's budget, with the token as the unit of account inside that exchange.
- Evidence
- Primary paper
- Re-measured
- No
- Assumptions
- 4
- Records linked
- 3
Code
Source, repository, and what is actually hosted
- BAT white paperhttps://basicattentiontoken.org/BasicAttentionTokenWhitePaper-4.pdf
- The source URL is not a GitHub repository. No code remote is invented.
- Full text is not mirrored. Public availability is not a licence to copy the file onto this site.
- Add to a reading list
01 Claim ledger
What the paper is allowed to say
Each row is a sentence already in the study. The status is the same on every row: a model claim, not a live measurement.
Claim 01 · Paper model
The problem named is the advertising stack
The paper argues that intermediaries take the surplus, users are tracked, and publishers are underpaid. The proposed fix is a browser that can see attention locally and settle the result.
Claim 02 · Paper model
Attention is measured on the device
The browser is supposed to decide what a user actually looked at. That measurement is the data the payment depends on. It is also a new privileged position for the browser maker.
Claim 03 · Paper model
The token clears three parties
Advertisers buy attention. Publishers and users are paid. The paper's flow uses BAT as the unit. A user does not have to understand a chain for the browser to make the transfer.
Claim 04 · Paper model
Privacy is a claim about local measurement
If the attention score is computed on the device and only a settlement moves, the paper can argue that a profile was not sold. If raw browsing is uploaded, the argument is gone. The implementation detail is the privacy claim.
Claim 05 · Paper model
This is not a general payments paper
The design is about one marketplace inside one user agent. Citing it as a blueprint for arbitrary creator tokens overreads it.
02 Three cuts
Observation, model, falsifier
A desk does not stop at the summary. Each claim is cut three ways, using only this study's own assumptions and checks. Nothing here is a new figure.
01 The problem named is the advertising stack
Observation
What the study says
The paper argues that intermediaries take the surplus, users are tracked, and publishers are underpaid. The proposed fix is a browser that can see attention locally and settle the result.
Model
What has to hold
The client actually measures locally what the paper says it measures.
Falsifier
What would retire it
Where is attention computed, on the device or on a server?
02 Attention is measured on the device
Observation
What the study says
The browser is supposed to decide what a user actually looked at. That measurement is the data the payment depends on. It is also a new privileged position for the browser maker.
Model
What has to hold
The client actually measures locally what the paper says it measures.
Falsifier
What would retire it
Where is attention computed, on the device or on a server?
03 The token clears three parties
Observation
What the study says
Advertisers buy attention. Publishers and users are paid. The paper's flow uses BAT as the unit. A user does not have to understand a chain for the browser to make the transfer.
Model
What has to hold
Users and publishers are willing to use that client. Distribution is the constraint.
Falsifier
What would retire it
Where is attention computed, on the device or on a server?
04 Privacy is a claim about local measurement
Observation
What the study says
If the attention score is computed on the device and only a settlement moves, the paper can argue that a profile was not sold. If raw browsing is uploaded, the argument is gone. The implementation detail is the privacy claim.
Model
What has to hold
Settlement batches do not reveal more browsing history than the privacy claim allows.
Falsifier
What would retire it
Where is attention computed, on the device or on a server?
05 This is not a general payments paper
Observation
What the study says
The design is about one marketplace inside one user agent. Citing it as a blueprint for arbitrary creator tokens overreads it.
Model
What has to hold
The client actually measures locally what the paper says it measures.
03 Sequence
One action, as an operating tape
- 01An advertiser locks a budget denominated in the token.
- 02A user browses with a client that records, locally in the paper's design, which publisher surfaces received attention.
- 03The client computes a payment allocation from those measurements and the campaign rules.
- 04A settlement transaction, often batched, pays publishers and the user's share.
- 05Verification, in the paper, is about whether the client followed the measurement rules, not about replaying every page view on a public chain.
04 Load-bearing
The argument, and where a pitch drops it
The browser becomes the exchange
The cut
Whoever ships the client defines attention, decides which sites qualify, and can change the allocator. That is a great deal of power even if settlement is on a public chain. The paper moves trust from ad networks to the client. It should be described as a move, not as a deletion.
Why it carries weight
If this cut is skipped, the paper's name is being used without the mechanism that makes the name mean anything.
Where it is dropped
An attention metric is a product choice. It is not a scientific unit, and the paper should not be cited as if it were.
Attention is a constructed metric
The cut
Seconds in view, clicks, and scroll depth are choices. A payment system that calls its choice 'basic attention' has picked a formula. Publishers will optimise for the formula. The paper needs that formula to be stable and public enough to be criticised.
Why it carries weight
If this cut is skipped, the paper's name is being used without the mechanism that makes the name mean anything.
Where it is dropped
An attention metric is a product choice. It is not a scientific unit, and the paper should not be cited as if it were.
A token does not fix an ad market by existing
The cut
Advertisers show up if the audience is real. Publishers show up if the payout beats the alternatives. The mechanism can be correct and the market still thin. The paper is a clearing design. Traction is not a theorem in it.
Why it carries weight
If this cut is skipped, the paper's name is being used without the mechanism that makes the name mean anything.
Where it is dropped
An attention metric is a product choice. It is not a scientific unit, and the paper should not be cited as if it were.
05 Register
What has to be true
Model · Not re-measured
The client actually measures locally what the paper says it measures.
Data · Not re-measured
Users and publishers are willing to use that client. Distribution is the constraint.
Model · Not re-measured
Settlement batches do not reveal more browsing history than the privacy claim allows.
Data · Not re-measured
The metric of attention is specified well enough that a publisher knows what they are selling.
06 Divergence
What happened after the paper
Brave shipped a browser and a BAT rewards system based on this thesis, with custodial options and later changes to how rewards work. Those product choices, especially anything custodial or KYC-gated, have to be compared with the paper rather than assumed to match it. The paper is the attention-clearing idea.
A later client, parameter set, or reward formula is a different object. Cite this paper for the mechanism. Cite a primary release for the network. This desk has not re-run the proof.
07 Pre-mortem
What to check before you use the idea
- 0 of 4 marked on this browser. A mark is a reading note, not a pass, a rating, or a recommendation.
08 Anatomy
The paper, in the order a builder needs
The problem it names
Web advertising in the paper's account is inefficient because intermediaries sit between the advertiser and the person, and because tracking is the measurement tool. Brave proposes that the browser, which already knows the page is on screen, can be the meter.
What the design proposes
- Users, publishers and advertisers are the three roles.
- Attention is measured locally. The paper's privacy claim depends on that locality.
- Payments are batched. The chain is not asked to settle every impression.
How the mechanism is specified
- The browser is the trusted measurer in the design. A user who does not run that client is not inside the mechanism.
- Fraud control — fake attention — is acknowledged as the adversary. A token does not define it away.
- The paper is a product design tied to Brave. It is not a general standard for all browsers.
What this page does not treat as proven
- Local measurement is only private if the client behaves as specified.
- An attention metric is a product choice. It is not a scientific unit, and the paper should not be cited as if it were.
- We do not describe token markets or user income.
Why the desk still reads it
The relevant studio question is who holds the meter. BAT's answer is the user's client, with settlement batched. A venture that meters attention on its own server and then says 'blockchain advertising' has not adopted the paper.
09 Lexicon
Terms, opened into the record
- Attention
- The paper's metric of user focus on a publisher surface. It is a formula, not a raw fact.
- User agent
- The browser that measures attention and is therefore in a privileged position.
- Settlement
- The payment that moves after measurement. It is not a record of every page view.
- Campaign budget
- The advertiser's locked spend, which the allocator divides according to measured attention.
10 Repository
Every linked record on this page
Underlined words open a page that already exists: a concept, a protocol profile, a failure record, or another paper. If a word is not underlined, this desk does not have a record for it.
Concepts
Papers
This is Blockchain Lab's reading of a public design paper. It is not the paper, not a copy of it, and not an offer of tokens, equity, custody or a partnership. Later network behaviour can diverge from the text. Nothing here is investment, legal or technical advice.
Research status: Design paper. Last reviewed: 1 October 2026. This is a reading of a public paper, not investment, legal or security advice.