AllDigital Specialty built an AI-first insurer. Here's what it learned.
AllDigital Specialty Insurance operates differently from most carriers. When CEO Athula Alwis and his co-founders launched the company, they built it around machine learning from the start-no legacy systems to untangle, no workflows to redesign, no employees resisting change. Today, roughly 70% of the company's business is handled autonomously by AI systems, a figure that established insurers are still chasing.
"We started AI-first, and our architecture was always set up that way," Alwis said. "The idea was to give agency to the AI system. We did not have to deal with legacy data issues or the employee resistance you see in certain places."
How 70% autonomous actually works
The breakdown is straightforward: 30 to 40% of incoming business receives an automated approval from AI systems. Another 30% is declined outright. The remaining 30% goes to human review.
AllDigital Specialty frames this within four stages of AI maturity-recommendation, assistance, execution, and orchestration. The company currently operates at stage three, where machine learning models make decisions independently.
Recently, the company deployed agentic AI to manage submission intake, clearance, and pre-analysis preparation. These tasks previously created friction for clients and internal teams. The result is faster processing across the entire workflow.
But Alwis is clear about one non-negotiable element: human oversight remains mandatory. "We need human governance in everything we do," he said. "Model governance, changes, and guardrails are all monitored by human experts." In a regulated industry, that oversight is foundational, not optional.
The team behind the systems
AllDigital Specialty's AI capabilities didn't come from vendor contracts or rushed hiring. Two of the six co-founders are AI specialists with decades of experience, including one with a PhD in AI research. When the founding CTO retired last November, his replacement held a PhD in AI applications and immediately advanced the firm's agentic capabilities.
The current team includes two junior AI engineers in their second year and a hybrid software-and-AI engineer. The company deliberately invests in mentorship and in-house training rather than simply buying senior talent off the market.
Domain expertise as design principle
Alwis describes the approach as "human-in-the-loop training"-pairing data-driven models with subject matter expertise to keep outputs grounded in insurance reality. "Domain expertise is one of our key design principles," he said.
The distinction matters. "People have come into this space with good technology but lacking domain expertise," Alwis said. "Because we are in the claims-paying business, at some point you write business and you have to pay claims. If you miss that point, you are going to learn the lesson the hard way."
Why AllDigital builds everything in-house
The company develops all AI systems internally. AllDigital Specialty holds a U.S. patent for training machine learning systems in the specialty insurance sector-a deliberate choice to protect intellectual property.
"Our AI IP is completely in-house with no outside involvement," Alwis said. "Infrastructure and other needs are handled through vendor partners within our predetermined guidelines."
When evaluating outside partners, the company applies a three-part test: Can they deliver on time, on budget, and within AllDigital's standards? If not, the company moves quickly to find alternatives.
What established carriers should take from this
For large insurers attempting to integrate AI, Alwis points to a recurring failure: bringing in expensive external talent without grounding them in insurance operations. "About 10 years ago, companies brought in very expensive data experts who did not know how the loss ratio is calculated," he said. "They had to start from scratch, it took a long time, and those efforts failed for the most part."
His recommended model for established carriers is hybrid: internal staff who understand legacy data definitions paired with specialists who can build modern machine learning systems. Critically, he argues for breaking down data silos-a cultural problem he identifies as one of the biggest barriers to successful AI for Insurance adoption.
"If I were running an insurance company, I would force a silo-less, frictionless environment where people work together and are rewarded for collective success," he said. "Even a simple machine learning tool can be misled by inconsistent labeling-someone calling something 'legal expenses,' someone else calling it 'ALAE,' when it could be the same thing."
The industry faces a talent shortage that complicates any strategy relying on external hiring alone. The deficit of qualified AI engineers in the U.S. has been estimated at as many as 80,000. AllDigital Specialty's response has been to grow talent internally, accepting a longer timeline in exchange for better cultural and domain alignment.
This approach mirrors broader conversations across the U.S. insurance industry about sustainable AI integration versus short-term technology plays. AllDigital Specialty's model-purpose-built, patent-protected, and persistently human-governed-represents what it looks like when AI Agents & Automation are embedded from inception rather than bolted onto existing operations.
Your membership also unlocks:
