The situation with enterprise software is quietly, but deeply confused, due to the rise of AI coding tools such as volts, replicas, and cursors.
These services overturn one of the most basic technical decisions companies make. Whether to purchase software from an external SaaS vendor or build it in-house. This once-clear distinction is blurred as AI reduces barriers to building custom software.
I figured out the size of this shift at a hackathon party in San Francisco. There, I met Netlify CEO Mathias Biilmann and startup security chief Mark Dorsi. They showed us a live slack feed with new apps deployed on Netlify's platform. Many of these were created using AI coding services similar to Vault. And most were internal tools that companies built for their own use.
This is a change in the ocean. Historically, the decision to buy a SaaS product stems from the high cost and complexity of building software internally. Skilled developers are expensive, and internal IT teams often grow thin. However, the development of AI-assisted software changes economics.
Tools such as Bolt empower the new classes of developers that Biilmann calls “AI-Native Developers” and what Bolt calls “software composers.” These are often non-traditional developers from business or operational roles, and now they can use AI to build apps and generate feature codes via English prompts. Training for these new builders takes years or months, and not years of years of computer science research.
As a result, there will be a huge influx of new developers. This new supply means that businesses can hire more coders for less money. This will help businesses tackle software projects that they previously didn't cut.
“That's where the build and bi equations began to change,” Billman told me. “With millions of new people who can build software, the barriers break down. There's a dramatic increase in what a single internal developer can build within a company. That's a radical change in the entire ecosystem than people think.”
Where this shift first occurs
Over 10,000 new websites have been created through new AI coding tools and are launched daily on Netlify's platform (coded only by human developers).
Biilman is looking at the first build and bi shifts happening in these software realms.
- HR, Training, Q&A: These are often very simple applications that read and write data and visualize it.
- Revenue Management, CPQ, Business Dashboard: AI coding tools and agents can easily build interfaces or visualizations in addition to existing corporate data.
- Marketing Tools: These are relatively simple from a software perspective, and often require highly customization to suit the needs of a particular company. What happens if the AI agent creates these applications?
“It's difficult to think of the full meaning because there are many other areas that will be affected by this,” added Billman. “But these are some of the simple, drooping areas.”
Who is building what?
Netlify has built an internal employee survey tool using AI coding tools. This is a task that is usually outsourced to SaaS providers such as Qualtrics or Momentive.
Another example: Netlify revenue operations staff could use Volt to create price calculators for enterprise transactions, potentially replacing the category of SAAS known as CPQ (Configure, Price, Quote) software.
Additionally, Netlify recruiters have built a new in-house interview training course app to hire managers using an AI coding tool called Lovable instead of purchasing from external providers.
Even VCs embrace this trend by actually doing some of the work themselves. Martin Casado, general partner at Andreessen Horowitz, recently built his own AI-powered customer relationship management tool. CRM software is the core offering of high-tech giant Salesforce.
“Some software makes it faster to code your version with AI than to learn someone else's non-intuitive shit UI,” Casado wrote in X recently.
Casado's CRM tool syncs with his calendar and email accounts. They need that information, plan an AI model weekly and investigate the people and businesses he plans to meet.
Notes on note
Martín Migoya, CEO of Tech Consulting Firm Global, discovered this trend.
“We're looking at projects that come out like that,” he said in a recent interview that some companies are trying to replace inadequate software services with internally built alternatives.
Still, he urged his attention. “It's difficult to create an enterprise-class product,” added Migoya. The AI model has improved, but the software will need support, the CEO said.
His views reflect general concerns: maintenance and reliability. If something breaks in an AI-coded internal app, who will fix it? This is where infrastructure plays a key role.
To address these concerns, companies like Netlify are developing what Biilmann calls the “opinion platform.” This is an infrastructure stack optimized for AI-generated code and agents. These platforms handle authentication, authorization, staging, security, and data access in a standardized way, reducing operational burden.
This evolution could allow companies to move from building prototypes to deploying production-grade software without hiring an armed force of senior support engineers, he added.
Strategic Implications for SaaS
The impact on the SaaS industry is important. Traditional SaaS models can be threatened if a company can build tailored internal tools at the same cost as a SAAS subscription and at the same cost as: Pricing per seat can look expensive compared to more abundant AI assist, in-house human coders.
Salesforce, the standard bearer of SaaS, can become vulnerable. The core CRM Recording System may remain sticky, but some of the custom features layered above may be replicated by AI coding tools.
“The Salesforce is really scary,” Billman said. “They don't have a coded atmosphere away from being a system of recording all sales data. But with that, many of their own custom features, which add to them, are far more likely to be replaced by people who build their own custom features.”
So far, Salesforce's answer is to build and launch your own AI agent. This Agent Force business shows early signs of traction with its customers, so this company may be good at dodging threats.
Others may not be that lucky. Consulting firm Alixpartners recently warned that over 100 mid-market software companies are stuck in the middle of this disruptive AI trend.
Large new software layers
A new wave of this software will likely not replace the system of records, but will be placed on top of it. Think internal dashboards, pricing calculators, contract tools, HR research apps, and media surveillance systems. All of this is traditionally handled by SaaS vendors. But once internal developers acquire a new AI superpower, building these tools in-house becomes a viable option.
Help us improve our business, tech and innovation coverage at BI by sharing a little about your role. This will help you coordinate the content that matters most to people like you.
What is your position?
(2 of 1)
What products or services do I approve purchases in my role?
(2 of 2)
Continued
By providing this information, Business Insider agrees that this data can be used to improve the site experience and use it for targeted advertising. By continuing, you agree to accept the terms and privacy policy.
Thank you for sharing your insight into your role.
AI-assisted software development is scaled to the scale of build and billing discussions. It won't kill the SaaS industry overnight, but we'll show you a powerful alternative.
Companies that embrace this model may be able to gain more control, reduce costs and innovate faster. Those who cling to traditional Thirs may end up paying more for less while their competitors build their future from within.
“You can become a software developer without writing any code,” says Biilmann. “It will have a ripple effect on everything that's been built.”
Sign up for BI's Tech Memo Newsletter here. Please contact me by email abarr@busienssinsider.com.
