AI Could Redraw DeFi’s Competitive Moats, Applebaum Argues

By Elena Vasquez

In a Nov. 10 essay, Multicoin Capital general partner and venture co-head Spencer Applebaum argued that artificial intelligence is lowering DeFi software costs and shifting competition toward customer access, capital and proprietary underwriting data.

His thesis goes beyond faster coding. Applebaum expects cheaper protocol development to weaken the developer-ecosystem advantage of general-purpose Layer 1 and Layer 2 networks, placing greater weight on firms that retain users, absorb losses and assess risk. He offered no benchmark or timetable, leaving a strategic framework rather than a measured forecast.

Which advantages matter when code gets cheaper?

Applebaum organizes his case around three assets that software alone cannot readily reproduce:

• Distribution: Lasting customer relationships generate order flow and keep users connected to a financial product. He described distribution as “more than making an app available,” arguing that an established channel can be harder to replicate than code.
• Capital: A balance sheet can absorb losses. Applebaum cited risk-bearing prime brokers and reserves held by protocols as examples of capital that can support a business through losses.
• Proprietary underwriting data: Specialized information used to assess risk before extending credit or insurance can shape those assessments. He pointed to OnRe reinsurance and USD.ai GPU-backed loans as examples.

The three advantages address separate operating constraints. Distribution determines whether a product can retain activity; capital determines how much loss it can carry; underwriting data determines whether a firm can assess collateral and risk credibly. Reinsurance and asset-backed lending depend on judgments about particular risks and collateral, making information part of the product itself.

Multicoin general partner Shayon Sengupta has also proposed a market for financing GPU and compute claims, including claims tied to GPUs, future computing capacity and customer offtake receivables. The proposal applies the same data-and-capital framework to compute financing, but it comes without a stated market size or implementation timetable.

In plain terms, cheaper code raises the value of access, loss capacity and proprietary risk information.

Can risk tolerance create a defensible startup?

Applebaum sees risk tolerance as a possible entry point for startups. They can target areas established firms avoid because of compliance or reputational concerns, then try to build lasting advantages around customer relationships, capital and data. In his formulation, “taking on risk creates an opening”; it does not guarantee a durable business.

The sequence matters because initial risk acceptance is not itself a moat. A startup still must develop one or more of the three advantages, and Applebaum’s account identifies no specific company or evidence that willingness to bear risk alone creates a durable position.

In plain terms, risk can open a market, but access, capital or data must sustain the position.

What changes for general-purpose blockchains?

Applebaum described separate pressures on general-purpose blockchains and developer ecosystems. He said general-purpose Layer 1 and Layer 2 networks are losing distribution and capital advantages as interoperability improves and users and capital reach DeFi through other channels. Cheaper software chiefly weakens the developer-ecosystem advantage by lowering the cost of building protocols.

The distinction is important. Interoperability can reduce a chain’s reliance on its native user base, while cheaper protocol development reduces the cost of competing through code alone. Together, those shifts would favor firms that can bring users, fund losses and assess risk over those relying mainly on technical differentiation.

That conclusion remains conditional rather than measured. He identified no specific network, migration statistic or deadline for erosion in distribution or capital. If interoperability routes more users and capital through other channels, general-purpose networks’ competitive position could weaken further, though Applebaum provided no timetable for that market-structure change.