Author: Jon Reiter
Compiled by: Saoirse, Foresight News
In August 2023, we published an article arguing that Base was essentially just a tool for Coinbase to provide centralized custody and transfer services that bypassed customer identification and anti-money laundering checks, using obscure technical jargon and vague, futuristic slogans to deliberately obscure the true nature of this business. Recently, Jesse Pollak (co-founder of Base) publicly confirmed that the Base application has been reassigned to the parent company Coinbase for overall planning. The well-known industry figure Jordan Fish (alias Cobie) will independently manage this product line, and the product may even expand beyond the Base ecosystem in the future. This development makes our earlier viewpoint increasingly difficult to refute.
Coinbase holds numerous financial licenses globally. Such operations, both in the past and present, pose serious compliance issues. As early as February 2023, Coinbase voluntarily admitted that Base had absolutely no decentralized attributes at launch and even publicly released a decentralization roadmap that year. The details will be discussed later, but the conclusion is this: that roadmap was shelved before any substantial progress was made. The authoritative Layer 2 monitoring platform L2Beat currently rates Base at Stage 0 (this rating will be downgraded from Stage 1 to Stage 0 in August 2026), with the entire network still fully managed and guaranteed by the platform. Simply put in industry terms: this highly respected L2 monitoring institution judges Base to have no decentralization characteristics whatsoever.
After our August 2023 article was published, Coinbase promptly issued a decentralization commitment for the Superchain ecosystem. The document straightforwardly admitted that Base was fully operated by Coinbase and claimed it would collaborate with Optimism and the broader Superchain ecosystem to gradually address the centralization issue. However, like the previous roadmap, this commitment was also shelved without substantial progress.
In 2024, Vitalik publicly called for the industry to stop heavily promoting Stage 0 L2 products and to fully push all projects towards the Stage 2 decentralization standard. In February 2026, Vitalik directly challenged the core idea that "Layer 2 scaling is the optimal solution for Ethereum scaling," citing the reason that "progress towards the Stage 2 standard has been far slower and more difficult than initially envisioned."
It must be objectively stated: whether in 2023 or 2024, no project in the industry had presented a mature, technically sound Stage 2 solution that balanced security and scalability; to this day, the projects Vitalik mentioned still haven't completed this full solution. This isn't due to teams failing to execute a clear plan, but rather because the entire industry's R&D efforts have yet to overcome the core technical challenges. An analogy: it's like a Mars exploration program where the full rocket blueprints aren't even finished before the R&D funds are exhausted, forcing the project to be shelved. The industry's initial bet on rapidly achieving a decentralized technical path has now failed.
The above is a concise summary of the industry's state over the past three years. The vast majority of Ethereum's Layer 2 networks face similar issues: progress towards decentralization for all L2 projects is either slow and intermittent or has completely stalled.
Next, we will meticulously dissect Base's complete development trajectory, then confront a core problem the industry avoids discussing: when a Layer 2 network remains stuck at a fully centrally controlled Stage 0 for an extended period, what difference is there between it and a custodian financial institution or a funds transfer service provider? When should regulators launch investigations into such platforms that lack proper licensing and bypass customer identification and anti-money laundering processes?
Base's Development History
Coinbase's initial roadmap planned to implement a permissionless fraud-proof mechanism by 2023. Even without understanding the technical details, the key point is: this core component was not launched as scheduled in 2023. It is a necessary element for removing the official control and backstop mechanisms, and its delay directly sowed significant risks.
It wasn't until April 2025 that Coinbase finally announced achieving this milestone. At that time, L2Beat rated Base as a Stage 1 Layer 2 network, defined as "essentially decentralized but still retaining an emergency backstop mechanism that officials can fully take over." Of course, stage classification involves some subjective judgment, and industry rating standards are constantly updated; the technical level achieved in April 2025 would likely be classified as a fully centrally controlled Stage 0 under today's newer standards. Notably, multiple security incidents in recent years have shown that the so-called "emergency-only" control permissions are being used with increasing frequency, with actual control far exceeding what project teams initially claimed, leading to a continuous tightening of industry stage classification standards.
This is a common disorder in the Web3 industry: not just superficial "decentralization theater," but project teams also repurposing backend control permissions originally intended for "security protection and user asset safeguarding" for other uses, and in extreme cases, even directly stealing user assets. Many protocols touted as decentralized have suffered massive losses due to teams secretly holding administrator keys that were later leaked, leading to asset theft.
For years, this publication has adhered to a conservative evaluation standard: all centralized control permissions should be analyzed under a worst-case scenario, assuming malicious intent by the operator, and strictly assessing "whether the platform has the ability to abscond with all user assets." Past security incidents have repeatedly validated the reasonableness of this evaluation logic, and now an increasing number of practitioners are beginning to share this view.
Returning to Base's main development thread: even with delays and limited effects, the project did achieve some progress outlined in the roadmap. However, in February 2026, Coinbase announced it was completely abandoning the original plan to advance decentralization via Optimism, opting instead to build a technology stack fully controlled by Base itself. Before this adjustment, some administrator permissions for Base were jointly held by Coinbase and Optimism; after the adjustment, all control permissions were consolidated into two multisignature wallets: a Centralized Security Council Multisig and a Centralized Sequencer Multisig. Coinbase, with its substantial capital and huge industry influence, could already exert significant intervention on Optimism, and now can completely sway the individuals and small partner institutions behind these two multisig wallets.
At this point, Coinbase has not only fallen far behind its 2023 plan but has also completely abandoned the entire roadmap without introducing any viable alternative development plan. More crucially, the small portion of control temporarily diverted to Optimism has been entirely reclaimed by Coinbase. Joint ventures could have co-custodied assets; the collaborative structure between Coinbase and Optimism briefly achieved this, representing at least a tiny step towards the 2023 roadmap. But with the roadmap completely discarded, Base can no longer provide any substantive basis for advancing decentralization. Marketing hype does not equal implementation; mere buzzwords don't count as actual development.
In June 2026, Base suffered two network outage incidents. During the repair process, Coinbase directly modified the underlying code, rolled back the entire blockchain's data, and forced all network nodes to synchronize a fix package to resume operation. After three and a half years of operation, Base remains at a fully centralized Stage 0, where the platform can use its control permissions to directly dispose of user assets at any time. The facts are clear.
A Centralized Distributed Database
The repair operations during the two outages completely exposed Base's essence: it is merely an ordinary centralized distributed database. There are countless mature commercial solutions on the market: Oracle, SAP, IBM all offer highly stable commercial databases; MySQL can be deployed for free; major cloud providers also offer cloud-based distributed database services. Distributed databases are a mature, implemented technology, stable and reliable as long as transaction volumes match hardware capacity limits, only encountering performance bottlenecks under extremely high traffic loads.
Base's transaction throughput is only a few hundred transactions per second (TPS), a performance level equivalent to traditional systems from the 1980s and 1990s. Reviewing industry reports from the dot-com bubble era 20-30 years ago reveals that many traditional enterprises back then handled transaction volumes orders of magnitude higher than Base. This performance level predates the birth of Bitcoin and even the widespread adoption of smartphones; early online banking was only accessible via computer web pages, requiring users to manually check HTTPS encryption indicators before entering passwords—a very antiquated technical level.
One can look up reports from 2001 about eBay's system failures in the late 1990s. We do not underestimate the challenges of system operations and have personally experienced the technical limitations of the 90s: back then, you could only read digital camera pictures via serial port; working with medical imaging, a computer worth as much as a high-end car couldn't load dozens of 16-megapixel黑白 (B&W) images at once; whereas today, the cheapest Samsung phone can easily store and smoothly open many times that number of image files.
But this doesn't mean Base is tackling antiquated technical challenges. Achieving 1990s-era traditional server performance in a permissionless, globally distributed, decentralized network might be extremely difficult, or perhaps not—but Base isn't even trying in that direction. Today's Base is simply an inefficient, unstable, poorly performing distributed database, a technology that was fully mature decades ago. Coinbase had grand visions for Base, and the industry in 2023-2024 could still be somewhat accommodating, debating whether regulatory standards should be relaxed for innovation; but by 2025, this rhetoric had become tiresome.
Base has been live for three and a half years, built on blockchain at its base, but this blockchain hasn't been used to create any new, original product logic. The entire system, from start to finish, remains centrally operated by Coinbase. The blockchain instead unnecessarily increases operational complexity, equivalent to choosing "hard mode" for running a long-mature database technology, while consistently failing to achieve decentralization through it. Regulators should not grant special leniency simply because a company deliberately chooses a more complex, troublesome technical solution.
Why We Consistently Emphasize the Entity is Coinbase, Not Base
In this article, we frequently attribute control to Coinbase rather than Base, a statement supported by ample evidence. One can review the Coinbase official announcement blog. The recruitment section at the end contains this passage:
Building the next generation of the internet is a collective effort. If you're passionate about scaling, security, or advancing Base's decentralization, we invite you to apply for our open positions. Click here to view opportunities.
Clicking the recruitment link redirects to the Greenhouse hiring platform, where all positions clearly state that hired personnel belong to Coinbase, as either full-time employees or outsourced contractors. In the announcement's comment section, Coinbase's official employees also uniformly refer to the Base team as "we." Base's website terms of service and privacy policy are also replete with references identifying Coinbase as the entity.
Coinbase's external narrative is that Coinbase merely acts as an incubator, and once the project achieves decentralization, Coinbase bears no related legal liability. However, the reality is clearly visible: the entire system is operated by Coinbase full-time employees; and a Stage 0 Layer 2 network possesses no decentralization whatsoever. Therefore, Base is essentially a Coinbase business unit. This is the core reason this article consistently names Coinbase directly.
Industry Prospects and Legal Controversies
A core question often raised about the Web3 industry is: many projects don't need blockchain at all but force-fit chain technology. Base's initial explanation was: blockchain is the only technology with the potential to efficiently solve scaling problems—a view that remains debated to this day. But after years of development with zero substantial progress, three pointed questions must be put on the table:
- Did Coinbase's decentralization roadmap, from the very beginning, lack a complete, actionable implementation plan?
- Given the stagnation of Base and the broader L2 industry, is achieving full decentralization feasible in the short term?
- How much tolerance period should regulators grant for innovation, and when should compliance investigations be initiated?
Coinbase holds financial licenses in multiple countries, legally prohibiting it from operating a funds transfer platform that bypasses customer identification and custodial regulations. This is both a hard obligation attached to various licenses and a law uniformly applicable to all market entities. It is reasonable for regulators to show some tolerance during a new project's initial launch phase; software development inherently has bugs, and user asset security needs time to mature—the industry understands this. But no reasonable argument suggests that a project can enjoy indefinite regulatory leniency after 40 months with zero progress towards decentralization.
As a publicly listed company on the US stock market, Coinbase has an obligation to release truthful, accurate public information, especially regarding its business plans, development prospects, and shareholder interests. If Coinbase never had a concrete decentralization implementation plan from the start, and only promoted vague "conceptual ideas" externally, serious integrity issues arise: Did the company deliberately conceal its technical shortcomings and falsely advertise to the market that it possessed decentralization implementation capabilities?
Years of development work with zero progress look less like industry building and more like deliberately consuming the regulatory grace period. From February 2023 to August 2026, Base's decentralization rating has remained at Stage 0, with no positive breakthroughs. The development team did complete coding, launch, and other foundational work, and the platform has processed substantial user assets, performing "work" in a physical sense; but the so-called output is only the team's continuous marketing hype, yielding no substantive results on the core goal of decentralization.
Even if Base adds more features and more complex underlying mechanisms, it's irrelevant. The evaluation criterion has never been whether engineers find the R&D interesting or whether basic development work is done. The core standard is: Can this system provide legal, valuable services that align with Coinbase's publicly stated project plans?
Based on the current situation, Coinbase's series of operations essentially attempt to package "we never mastered the path to compliant decentralization" as "we are tackling major industry challenges and deserve permanent exemption from regulatory constraints." This rhetoric was already heavily questioned years ago and is now full of holes.
Corporate innovation deserves a reasonable trial-and-error space; we don't demand that all project plans of a public company be perfectly executed without deviation. But years of consistently unmet plans, coupled with an operational model suspected of conducting unlicensed custody and funds transfer businesses, must bear corresponding legal consequences. Otherwise, financial regulatory rules would lose all binding force.
