What Happened
Token economics · Crypto · Governance · Policy BriefRoadmap 4.0 of Jul 2026 · record to Aug 2026
Executive summary · one-page brief
A Retreat Executed as a Pivot
Brave's fourth BAT roadmap moves the token from paying for attention to paying for things, and begins a gradual sunset of the Brave Ads notification unit — the use case BAT was sold to power in 2017 — with rewards continuing to accrue on it through the transition. What is new is a buyback promise whose percentage is never stated, routed into a pool Brave's own terms call company-owned and discretionary, on a transparency page that has listed no BAT purchase in 311 days. The retreat did not start this July; it started in January 2023 — and in the seven weeks since this roadmap was published, BAT has fallen to a new all-time low.
The Numbers That Frame It all from Brave's own pages or two independent market-data providers
- 311days since the last BAT purchase Brave has listed, as of 28 Aug 2026
- −81.4%BAT purchased, 2025 vs 2024 (rows Brave itself lists)
- $0.055BAT's new all-time low, set 14 Aug 2026 — below its 2017 low
- 122.6Mmonthly users, as of 31 Jul 2026 — the browser is working
- 1.81Mcreators Brave's own page counts, vs its “over 2 million” claim
What Roadmap 4.0 Actually Proposes seven initiatives, graded by how developed each one is
| Standing | Initiative | What's actually confirmed |
|---|---|---|
| Confirmed, unspecified | Rewards Card | A virtual and physical card on GENIUS-compliant stablecoins, paying BAT cashback — issuer, geography and rates are all deferred to a future announcement. |
| Confirmed, unspecified | BravePay | A stablecoin payment layer using .brave addresses and “private transactions” — merchant support and an SDK are described only as a longer-term goal. |
| Confirmed, deferred | Creator Contribution Protocol | Micro-royalties when AI products access registered creators' content — a dedicated post was promised “later in 2026”; none had appeared as of 28 Aug 2026. |
| Confirmed, running | x402 + MPP | HTTP 402 agentic payments, “equal first-class support” for two rival stacks — an experimental repository already exists. |
| Confirmed, unquantified | Buybacks | “A percentage of Brave's net revenue” funds BAT buybacks via the UGP Reserve — no percentage anywhere in the announcement, and the promised tokenomics post had not appeared as of 28 Aug 2026. |
| Confirmed, in progress | Brave Ads notification unit | Begins “a gradual sunset by the end of 2026” — rewards continue accruing on it during the transition; nothing indicates it has fully ended as of 28 Aug 2026. |
The Two Readings why “pivot or retreat” is a false binary
What is ending
The Founding Use Case
- Earning BAT required KYC from Jan 2023; virtual BAT was killed and unconnectable users lost balances.
- Users reported per-ad rewards falling 0.025 → 0.003 BAT (−88%), confirmed directly by a Brave VP on Brave's own forum.
- The User Share is only guaranteed to be “equal to or greater than” Brave's own share — which, if the two split the whole, floors it at roughly half, though the terms state no digit.
- Roadmap 4.0 itself does not end the unit outright: it begins “a gradual sunset by the end of 2026”, with rewards continuing to accrue through the transition.
What replaces it
An Unquantified Promise
- “A percentage” of net revenue — the figure is never given, in a post full of other figures, and the follow-up post promising one had not appeared as of 28 Aug 2026.
- Destination is the company-owned User Growth Pool, not a burn address, usable “in its discretion.”
- Language is “will fund”, “plans to”; the terms carry no buyback obligation.
- No purchase of any kind — advertiser-funded or revenue-funded — has appeared on Brave's own transparency page in 311 days.
Bottom line
Brave the company is succeeding; BAT the token is being repositioned for the fourth time. The replacement asks holders to trust an unquantified share of undisclosed revenue flowing into a discretionary pool — a claim about the future backed by less disclosure than the advertiser-funded programme it supplements, made while the token itself was, within five weeks of the announcement, trading below the low it set at launch in 2017.
Timeline
In Order the sunset is the last step, not the first
- 24 Apr 2019Brave Ads with user rewards launches for desktop — BAT's founding use case goes live, at an initial 70% user share.
- 13 Jan 2023Brave announces that a KYC'd custodial account becomes mandatory to earn, and that virtual BAT will be sunset.
- 9 Mar 2023A forum thread reports an 88% cut in per-ad rewards; a Brave VP confirms, in the same thread, that rewards float with advertiser bids and BAT's own price.
- “at the earliest… Apr 2023”Virtual BAT (vBAT) sunset — Brave's own post hedges the date rather than naming one. Users in countries with no custodial provider — Brave's own example is Germany — lose remaining balances.
- 27 Nov 2024BAT Roadmap 3.0, “The On-Chain Era”, is published.
- 21 Oct 2025The last BAT purchase Brave has listed anywhere. 159,174 BAT via Gemini. Nothing since, as of 28 Aug 2026.
- 26 Nov 2025The current BAT Rewards Terms of Service — the version defining the User Growth Pool and the User Share wording — is last updated.
- 9 Jul 2026BAT Roadmap 4.0 is published; the notification ad unit's gradual sunset by the end of 2026 is announced.
- 14 Aug 2026BAT sets a new all-time low, $0.0552–0.0553, below its 2017 low — per CoinGecko and CoinMarketCap independently.
- 28 Aug 2026Brave's transparency page still lists no purchase since 21 Oct 2025 — 311 days — while its browser-growth panel stands updated to 31 Jul 2026.
The purchase record is Brave's own, and it remains the strongest single data point. Reading only the rows the transparency page itself lists: 11 purchases totalling 4,606,715 BAT in 2024, 6 totalling 858,096 in 2025, and none in 2026. The last is dated 21 October 2025 — 311 days before 28 Aug 2026, and 283 days before the page's own “as of Jul 31, 2026” stamp — so the page is being actively maintained while this one table has not moved in over ten months. Brave is asking holders to trust a new, revenue-funded buyback mechanism at the moment its existing, advertiser-funded purchase programme appears to have stopped entirely. A structural absence matters just as much: Brave has never published BAT paid to users, BAT paid to creators, unclaimed BAT, or the User Growth Pool's balance — so nobody outside the company can compute whether either the old attention model or the new transactional one is actually working.
- 1,806,384creators counted directly from Brave's own per-platform table, vs its “over 2 million” claim
- $0revenue, funding, headcount or UGP balance Brave has ever disclosed
The Argument
The Destination Is Weaker three checks against the documents
| Claim | As presented | What the documents show |
|---|---|---|
| Buybacks | Revenue-funded BAT buybacks supporting rewards and growth. | No percentage stated; “net revenue” undefined for a private company; destination is a pool the terms call company-owned and usable “in its discretion” — a balance-sheet transfer, not a burn. No obligation in the terms, and nothing disclosed as bought under this mechanism as of 28 Aug 2026. |
| Agentic payments | x402 and MPP support; a growing space. | Two dated snapshots tell different parts of the same story. On 11 Mar 2026, reporting put x402 at roughly $28,000/day (~$0.20/transaction), calling it “mostly a mirage” with much of the volume gamed. By 15 Jul 2026 — six days after this roadmap — the protocol's governance had moved to a 40-member foundation including Visa, Mastercard, American Express, Stripe, Google, Amazon Web Services and Cloudflare, volume grown to about $24M across 75M payments in the prior month — still, by that same reporting's own comparison, a fraction of what Visa alone moves in a single day (~$40bn). |
| Privacy | Preserving “the anonymity … of cash in your back pocket”; “private transactions”. | GENIUS-compliant stablecoins must by statute maintain the technical capability to be seized, frozen or burned on a lawful order, and a card needs a compliant KYC issuer. The roadmap does not address this tension anywhere in its text. |
The Creator Protocol Is a Late Entrant same HTTP 402 mechanism, wrong layer
| Scheme | Since | Where it sits |
|---|---|---|
| Cloudflare pay-per-crawl | 2025; shifted 1 Jul 2026 toward pay-per-use | The CDN edge, across over 20% of web domains (Cloudflare's own figure). HTTP 402; new defaults blocking unpaid AI training/agent bots on ad-supported pages take effect 15 Sep 2026. |
| RSL | Launched 10 Sep 2025; reached its 1.0 spec 10 Dec 2025 | Server-side licence terms, launched with Reddit, Yahoo, People Inc., Internet Brands, Ziff Davis, Fastly, Quora, O'Reilly Media and Medium — supporting pay-per-crawl and pay-per-inference from the start. |
| Brave Creator Contribution Protocol | Announced 9 Jul 2026 | The browser — but AI training crawlers hit servers and CDNs, not browsers, and the roadmap does not explain how a browser-side protocol bills them. Unspecified and unbuilt. Brave's real edge is its 1.81M registered creators. |
What Others Add
Two Things True at Once the company's own numbers, checked against the token's
Two things are true at once, and they are frequently conflated. The company side is strong: 122.6M monthly and 51.7M daily users — a 42.2% ratio most consumer software would envy, an independent search index estimated at roughly 1.6bn queries a month, and revenue third parties estimate grew from about $52M in 2024 to roughly $100M in annualised run rate by 2025. (Revenue, funding and headcount are third-party estimates; Brave discloses none of them.) Roadmap 4.0 is not a rescue. The token side is roughly nine years old and, as of 28 Aug 2026, worse off than when it started: about $0.069, a market cap of roughly $103M, about 96.4% below its November 2021 high — and, since 14 August 2026, below the low BAT set at its own 2017 launch. Nothing in Roadmap 4.0's own text addresses that timing; the two events are five weeks apart, and nothing in either company or market data shows one causing the other.
What the Roadmap Stopped Mentioning and what was never independently checked at all
Roadmap 4.0 is the fourth in a lineage that has been accelerating: Roadmap 2.0 (22 Feb 2021) to Roadmap 3.0 (27 Nov 2024) is 3.8 years; Roadmap 3.0 to Roadmap 4.0 (9 Jul 2026) is 1.6 years. A first roadmap is reported, in secondary retrospectives rather than at Brave's own now-missing page for it, to have run from around June 2017 — which would put the 1.0-to-2.0 gap at about 3.7 years and BAT's token itself at roughly nine years old. Roadmap 4.0 also silently drops “The On-Chain Era” — Roadmap 3.0's own headline nineteen months earlier — mentioning neither Boomerang, nor Themis, nor disintermediation anywhere in its text. Items here are not cancelled; they simply stop being mentioned, which is the most useful thing to know when reading a fourth roadmap. One caveat on method: no substantive published criticism of Roadmap 4.0 itself appears among the trade press that covered it — coverage largely restates Brave's own post, and a general search of Brave's own community forum for reaction to the roadmap was not attempted. That is “not found”, not “does not exist”.
- $1.92 → $0.055BAT, Nov 2021 high to its new low, 14 Aug 2026
- 4throadmap in ~9 years; ~3.7y, 3.8y then 1.6y apart
Conclusion
So What the retreat continues; watch for the number
Where it lands
The retreat that began in January 2023 has not reversed; Roadmap 4.0 extends it, adding a genuinely new mechanism — a revenue-funded buyback — that is, on its own published terms, less disclosed than the advertiser-funded programme it supplements. None of this proves the buyback will not happen; it proves only that, as published, there is no percentage, no obligation, and, as of 28 Aug 2026, no transaction to point to.
Reframe
“Company thriving, token retreating” is not new to this roadmap; it has been the pattern since at least the 2023 reward cuts. What Roadmap 4.0 changes is the story told about it — from paying users for attention to promising them an unquantified share of a discretionary pool — while the one number that would let anyone check the promise stays unpublished.
- Watch for one disclosed buyback — a date and an amount on the transparency page is the cheapest test that Roadmap 4.0's central promise is real.
- Watch whether the promised tokenomics post ever names a percentage. Without a number, the promise stays unfalsifiable.
- Watch whether any bought-back BAT is retired or re-spent. A burn address would change the analysis; the company-owned User Growth Pool would not.
- Watch for a card issuer and a geography, and whether the notification unit's “gradual sunset” actually completes by the end of 2026 as stated.