The “SaaS apocalypse” is better understood as a possible reshuffling of software value than the end of subscriptions. AI may make it cheaper to build narrow, custom applications, giving open-source projects a chance to serve needs that once could not justify a standalone product. But cheaper creation does not solve the harder problems: maintaining software, securing it, supporting users, and funding the work over time.
What does “SaaS apocalypse” mean?
“SaaS apocalypse” is a market narrative: the idea that AI could weaken the economics of some software-as-a-service businesses by making their features easier to reproduce or replace. It is not evidence that SaaS as a whole is disappearing. In a March 10, 2026 analysis, BBVA Global Markets Strategy describes the potential disruption as selective, with risk varying by what a product does and how much value is embedded beyond its visible features.
BBVA identifies four concerns behind the narrative: AI platform commoditisation, stronger start-up competition, bespoke enterprise applications, and AI-driven seat compression. These are pressures investors and companies are watching, not measured proof that each outcome is occurring across the market.
Why could cheaper software creation help open source?
The opportunity begins with development economics. If AI tools reduce the time or cost required to build and adapt software, a small team may be able to attempt applications for specialized users who are too few to support a conventional commercial product. A tool for a particular profession, workflow, or community could become feasible even when its audience is too narrow for a large vendor’s roadmap.
#1 Best Overall
That is a plausible mechanism, not a demonstrated count of newly viable projects. The HackerNoon article that supplies the thesis predicts more software creation and a shift in where revenue accrues; those remain forecasts rather than established market outcomes. Open source could benefit because people can inspect, adapt, and share a common codebase instead of waiting for a proprietary vendor to prioritize a niche request.
Lower build costs also make competition easier. A small business might commission or assemble a tailored workflow rather than buy a broad subscription. A community may adapt an existing open-source project to a local need. In either case, the advantage is not simply that code is free: it is that the software can be changed to fit a problem that a general product handles imperfectly.
Which SaaS products are most exposed—and which may be more durable?
BBVA’s analysis suggests comparing products by how reproducible their core function is, whether they control authoritative records, and how costly or risky it is to switch. Its judgments are a market framework, not a universal classification; individual products can differ substantially.
| Software type | BBVA’s assessment | Why the distinction matters |
|---|---|---|
| Simple analytics, service desk, basic reporting, and single-feature marketing tools | More exposed to automation or in-house replication, according to BBVA Global Markets Strategy (March 10, 2026) | If the product’s main value is a relatively narrow function, customers may find it easier to reproduce, replace, or fold into a broader AI-enabled platform. |
| Systems of record, ERP, core databases, data security, and stateful infrastructure | Comparatively defensible, according to BBVA Global Markets Strategy (March 10, 2026) | Authoritative data, accumulated business logic, and switching friction can make replacement difficult, especially where errors or unauthorized changes carry high costs. |
This distinction helps explain why software creation can get easier without every established vendor becoming obsolete. A feature may be replicable while the customer’s trusted data, permissions, integrations, audit history, and operational processes remain difficult to move. The product that stores and governs those records may therefore be more durable than a tool that mainly transforms or displays them.
Recommended Free Tools
What do AI spending plans tell us—and what don’t they prove?
Rothschild & Co’s February 2026 Growth Equity Update reported that Microsoft, Meta, Alphabet, and Amazon planned about $650 billion in AI capital expenditure for 2026, compared with about $380 billion in 2025. These are reported plans, not independently confirmed audited spending totals. The scale signals heavy investment in AI infrastructure; it does not prove that SaaS is collapsing or that open-source projects will capture the resulting value.
For software businesses, the more immediate questions are whether AI changes customer expectations, reduces demand for paid seats, shifts pricing power, or lets new competitors build alternatives. A vendor that adds AI features must still show that those features solve a customer problem and support a viable business. Announcing AI capability alone does not establish that value.
Can open source turn the opportunity into lasting value?
Making an application is only the first part of delivering software. Someone must review changes, fix vulnerabilities, update dependencies, handle compatibility, provide documentation, and respond when users encounter problems. If a project becomes important to a business workflow, users may also need dependable releases, backups, access controls, and support.
Open-source availability does not by itself guarantee security, quality, funding, or resilience. Nor does it settle who pays for the labor that keeps a project usable. The sources informing this market debate do not establish that open-source maintainers will receive sustainable funding as software creation gets cheaper. That is a central uncertainty, not a detail to assume away.
Best Value
- Maintenance: A quick prototype can become a long-lived dependency. The project needs people responsible for bug fixes, upgrades, and compatibility.
- Trust and security: Visible code can enable inspection, but security still depends on sound design, timely fixes, and responsible deployment.
- Data and operations: Users need to decide where data lives, who controls access, and who is accountable for backups and service availability.
- Support and funding: A project needs a realistic way to pay for ongoing work, whether through services, sponsorship, paid hosting, or another model. No single model is guaranteed by open-source licensing.
- Distribution: A useful tool must reach the people who need it. Discoverability, integrations, documentation, and confidence can matter as much as the cost of writing the initial code.
How to judge whether a niche open-source tool is a real opportunity
For a team considering a new project—or a company deciding whether to rely on one—the key question is not just “Can AI build it?” A more useful assessment looks at the full lifecycle and the value that remains hard to copy.
- Define the underserved workflow. Identify who has the problem, how they handle it now, and why existing products do not fit. A narrow audience is an opportunity only if the need is real and persistent.
- Separate the easy-to-copy feature from the hard-to-replace value. Ask whether the application depends on unique records, specialized business rules, integrations, or a trusted operational history—or whether its core function can be replicated quickly.
- Plan for consequences of failure. Consider the cost of errors, outages, data loss, and unauthorized changes. The more consequential the workflow, the more the project needs mature operational practices.
- Name the people and resources behind maintenance. Identify who will handle security updates, releases, user support, and compatibility after the initial build. If that work has no credible owner or funding path, cheap development has not made the product sustainable.
- Choose a value model that fits the users. Open-source code can be shared while paid services such as hosting, support, or customization fund some of the work. Whether any model succeeds depends on users’ needs and willingness to pay; openness alone does not settle the economics.
The opportunity is a change in what is feasible, not proof of an apocalypse
AI could expand the range of software worth attempting, especially for narrow needs that were previously too expensive to address. That creates room for open-source projects to serve overlooked users and for customers to demand more tailored tools. But the durable advantage will not come from producing code alone. It will depend on trusted data, dependable operations, ongoing maintenance, and a way to sustain the people doing that work.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
