Vitalik Buterin published a detailed post outlining the Ethereum (ETH) Foundation's specific conditions for supporting DeFi projects, rejecting what he called "gambleslop" in favor of permissionless, privacy-preserving, and open-source finance.
The post draws a clear distinction between DeFi the EF will actively back and on-chain finance it will not.
The statement is notable for explicitly naming oracle security as an unresolved systemic risk and for defining a concrete eligibility bar - the "walkaway test" - that supported protocols must meet.
The EF's Criteria
Buterin wrote that the Foundation is "not interested in supporting 'onchain finance' or even 'defi' indiscriminately." The qualifying criteria are specific: permissionless access, open-source code, privacy by design, minimal centralized chokepoints, and no unnecessary trusted third parties.
Protocols must also pass the walkaway test - defined as continuing to function reliably even if the original development team disappears or becomes compromised.
That test, which Buterin has previously applied to Ethereum's own protocol, now extends explicitly to DeFi applications.
Oracle Security as a Priority
Buterin named oracle security and decentralization as areas requiring immediate and sustained attention, writing that "there are a lot of skeletons in the closet here."
He called for the ecosystem to direct concerted focus at oracle infrastructure, describing it as a known systemic weakness that has not received sufficient scrutiny.
Oracle vulnerabilities have been a persistent attack vector in DeFi, with oracle manipulation exploits accounting for hundreds of millions in losses across protocols since 2020.
Read also: 21Shares Adds Spot Sui ETF To Nasdaq As Multiple Issuers Race To List SUI Products
What the EF Will and Won't Back
Buterin drew a line between foundational innovation - he cited AMMs as an example of genuinely new financial paradigms 0 and incremental or extractive products.
He specifically described "dopamine-maximizing gambleslop" as a category the Foundation has no interest in supporting.
On the constructive side, he listed privacy-preserving CDPs, AI-assisted formal verification, wallet-side security agents, and improvements to DeFi licensing and forkability as priority areas.
He acknowledged that Ethereum's permissionless nature means anyone can deploy insecure or centralization-dependent protocols - the EF's stance concerns only which projects it will actively support, not what can be built.
The post fits a broader pattern of Buterin's public statements in early 2026, which have consistently emphasized rebuilding self-sovereignty and privacy properties he says eroded during Ethereum's scaling-focused years.
Read next: Meta Eyes Stablecoin Integration For H2 2026, Eyes Stripe's Bridge As Likely Pilot Partner