娇色导航

As hype about AI agents reaches new heights, an emerging theory suggests that the groundbreaking AI tools will kill the long-running SaaS business model.

The claim isn’t particularly new, but , with people like Microsoft CEO Satya Nadella voicing this position. On earlier this year, he suggested AI agents will have “multi-repo CRUD” (create, read, update, and delete) functionality, effectively rendering underlying SaaS tools obsolete.

[ Related: ]

“The notion that business applications exist, that’s probably where they’ll all collapse, in the agent era, because if you think about it, they’re essentially CRUD databases with a bunch of business logic,” he said. “The business logic is all going to these agents, so they’re not going to discriminate between what the backend is. They’re going to update multiple databases, and all the logic will be in the AI tier.”

Nadella isn’t the only software expert sounding the alarm about the future of the SaaS market in the age of AI agents. A couple of months after his appearance on BG2, Greg Isenberg, CEO of internet portfolio company Late Checkout, laid out his predictions in a , saying he predicts an AI agent revolution within the next 18 months as AI moves from co-pilot functionality to being an autonomous operator.

“The dam breaks when someone can say, ‘Analyze our Q2 performance,’ rather than clicking through Tableau, or ‘Optimize our ad campaigns,’ instead of navigating Meta’s ad manager,” he writes. “The expertise previously bundled with the software gets unbundled by agents.”

Then, within three years, software becomes increasingly invisible, Isenberg adds. “The final phase happens when the agents bypass the human interfaces altogether,” he continues. “The value proposition of SaaS, bundling software, workflow, and expertise into user-friendly interfaces unravels completely. The interfaces were designed for humans, but agents don’t need them.”

SaaS still provides functionality

Some software experts agree, but others believe the narrative that agents will replace SaaS is too one-dimensional, in that while agents will largely eliminate the need for humans to interact with the GUI interfaces of SaaS products, the backend SaaS functionality will still be necessary.

Agents will likely replace parts of the user interface layer for routine interactions, says , founder and CEO of AI-powered game and video generation company RiseAngle. But the underlying SaaS platforms will remain essential to ensure reliability and governance checks.

Agents will also become an important interface layer to SaaS tools, he adds, but users lean to software tools they understand and what makes them feel productive.

“Even when multiple platforms offer similar core features, people gravitate toward the one that matches how they think and work,” Vahdat says. “That’s why UI design, dashboard layout, and user onboarding aren’t superficial features. They’re fundamental to why SaaS thrives.”

The idea that AI will kill the SaaS model is oversimplified, Vahdat adds. “SaaS tools aren’t defined purely by their CRUD functionality or business logic,” he says. “What sets them apart is the ecosystem around them — the UI and UX design, the workflows they guide users through, the ability to organize complex information visually, and the communities that grow around them.”

SaaS adds agents

There’s a bit of irony in Microsoft’s Nadella talking about AI agents eating the SaaS market, adds , founder and CEO of support automation company QueryPal.

“The funniest thing about Satya Nadella predicting that AI agents will collapse SaaS is that he’s simultaneously cramming agents into every Microsoft product he can find,” he says. “Excel now writes Python, Teams transcribes meetings and suggests action items, and Dynamics spawns autonomous agents like it’s running a digital ant colony.”

What’s really happening is that SaaS companies are racing to become agent platforms before agent companies can become trusted enterprise vendors, Nag adds. “It’s a land grab where the prize is controlling how workers interact with AI throughout their day,” he says. “The companies that already own your workflow have a head start because they know exactly what you do all day, down to your typos and coffee breaks.”

AI agents converging with SaaS tools will create a new service: “Software that watches you work and gradually takes over the boring parts,” Nag says.

But there’s a huge constraint unrelated to whether agents can do the work or if SaaS can integrate them, with the bottleneck being the creation of interfaces that allow one human to manage their own small digital workforce.

“Try using ChatGPT to manage five different tasks simultaneously and you’ll quickly find yourself drowning in conversation threads and disconnected context,” Nag says. “Now multiply that by every SaaS tool you use. The winners will be whoever figures out how to make controlling 50 specialized agents feel as simple as clicking through your inbox. This is a design problem disguised as a technology race.”

AI agents and SaaS tools will need each other to work well. “Agents need SaaS’s data, customer relationships, and compliance frameworks,” Nag adds. “SaaS needs agents to justify their valuations in a world where anyone can spin up custom software. They’ll merge and converge because they have to, creating hybrid platforms where the software does the work while humans do … what exactly? That’s the trillion-dollar question.”

UI goes away first

While some IT leaders raise doubts about AI agents killing the SaaS market, others are more open to the suggestion. The transition has already started, with agents erasing the SaaS interface layer first, says , cofounder and CEO at automated security vendor 0rcus.

“Any SaaS product that exists just to abstract database actions or business logic is already in decline,” he says. “Agents don’t need a GUI. They move through APIs, write to databases, and execute logic without human friction.”

The impact on the SaaS market will be a structural shift, resulting in a split SaaS market, where some vendors become irrelevant and others become infrastructure, by providing clean APIs and agent-friendly schemas, Adams predicts.

“Agents still rely on backend software, but they treat those systems as interchangeable,” he says. “They’re not loyal to brands, platforms, or legacy UX. The functional layer still matters, but the value is migrating toward orchestration, not navigation.”

But other experts doubt SaaS will disappear. AI agents aren’t likely to replace all SaaS, because foundational systems of record will still exist, says , COO at AI-drive spending management provider Tropic. Instead of interacting with those databases through a clunky UI, users will turn to AI agents.

“My belief is AI agents will 100% replace the work of the vast slew of SaaS point solutions that have grown over the last 10 years as SaaS ate the world,” Etkin says. “Today, users have a ton of SaaS fatigue, logging into many different applications to do their job.”

Interacting with various SaaS tools has become overwhelming, he adds, and AI agents create a better opportunity for users. “Rather than interacting with an application via a UI, an agent can be interacted with via chat or voice, the natural human modalities to get work done,” Etkin says.

Still, organizations will continue to need backend SaaS functionality, he adds. “What SaaS created was core infrastructure applications that will continue to be required for organizations to maintain data integrity, to cement and establish processes, and to undergird some of the key business infrastructure,” he adds. “I don’t envision a world where we have a bunch of AI agents that are just talking to each other.”