RPA Isn’t Dead

For a while, I thought UiPath’s turnaround depended almost exclusively on adopting LLMs, that LLMs were the miracle that arrived just in time to save them, and that RPA was a relic of the past.

Now I see clearly:
RPA is alive, healthy, expanding, and will continue to grow for the foreseeable future. It’s simply a technology that addresses a different part of the workflow.
And LLMs don’t kill RPA, they take it to its logical next evolution.

UiPath is a company that had an accident, ended up on the operating table, but had everything it needed to survive and thrive.
The market is deeply discounting UiPath, but you shouldn’t.

At this point, I believe UiPath is one of the most asymmetric opportunities I’ve found. Possibly as asymmetric as when I built my IREN position between $5 and $10 earlier this year.

Understanding UiPath’s Core Offering

UiPath’s foundational product is RPA, Robotic Process Automation.
At its core, RPA is simply software designed to automate repetitive, rule-based tasks.

The distinction is key:

RPA
Deterministic
Rule-based
Always produces the same result
Interacts with digital systems exactly like a human would
But does not understand semantic context

RPA is dumb, repetitive, but efficient, fast, cheap, and reliable.

AI Agents
Understand context and meaning
Can use tools, reason, plan, and escalate
Are non-deterministic
Powerful, flexible, but inherently less predictable

In the words of Evan Cohen, Director of Product at UiPath:

“Robots do. Agents think. Humans lead.”

And that’s the key.
UiPath is the company that figured out how to combine all three: robots, agents, and humans, into a single, coherent operating model.

Maestro: The Missing Piece of Enterprise AI

UiPath’s agentic orchestration platform, “Maestro” is the control plane that coordinates robots, agents, humans, and data across entire business workflows.

It is the missing layer in making agentic AI production-ready for the enterprise.
It transforms LLM-powered agents from experimental tools into governed, auditable, reproducible units of work, something enterprises desperately need.

With UiPath:
Businesses can design workflows that synergize robots + AI agents + human oversight
Deploy AI in a safe, governed, rule-based framework
Measure, track, and guarantee outcomes
Integrate directly with platforms like Microsoft Copilot, CrewAI, LangChain, and many others

It is the future of converging human and digital workforces.

UiPath Used to Automate With Its Hands Tied Behind Its Back

UiPath was wildly successful in business automation long before LLMs.
Gartner named them a Leader in the Magic Quadrant for RPA for seven consecutive years (including 2025).

But in hindsight, it’s clear:

Before LLMs, UiPath was trying to automate global enterprises with both hands tied behind its back.

AI is the unlock, the missing capability that allows RPA to transcend rules and take on unstructured work.

This evolution has reinvigorated the company.
The internal tone has shifted from survival mode to opportunity mode, and you can feel it across Dines’ recent interviews and quarterly results calls. UiPath has rediscovered direction, discipline, and urgency.

Valuation: The Growth Question

For my valuation, the central question is growth.

Yes, UiPath faces competition.
Yes, other players have agentic AI platforms.
And yes, Palantir arguably has an early lead in certain enterprise AI orchestration use cases.

But the automation market is expanding rapidly, and UiPath is:
Serving nearly 11,000 enterprise customers worldwide
Already generating $1.5B in annual revenue
Holding a healthy 83% gross margin
Entering a new product cycle with agentic AI
Sitting on a strong balance sheet
Running a more disciplined operating model post-restructuring

When modeling scenarios, I tried:
1. UiPath beats the market, possible, but uncertain
2. UiPath loses market share, extremely unlikely given their broad client base and deep understanding of enterprise processes.
3. UiPath grows with the market, most realistic

Market growth expectations across the best research I found are:
22% to 43% annual growth for the next 5–10 years
RPA is expanding
AI in automation is exploding
Enterprise orchestration demand is accelerating

Sources:
https://grandviewresearch.com/industry-analysis/robotic-process-automation-rpa-market
https://globenewswire.com/news-release/2025/08/18/3135131/0/en/Robotic-Process-Automation-Market-Size-to-Hit-USD-211-06-Billion-by-2034-Driving-Enterprise-Efficiency-and-Cost-Savings-Through-Automation.html
https://mordorintelligence.com/industry-reports/robotic-process-automation-market?
https://technavio.com/report/ai-in-rpa-market-industry-analysis?
https://market.us/report/intelligent-automation-market/#

Given these ranges, I selected 23.6% (from http://market.us) as my long-term growth assumption.

Expense Discipline & Operating Leverage.

After watching several interviews with Daniel Dines and reviewing the company’s restructuring history, it’s clear the leadership team has learned painful, but valuable, lessons:
Growth must be paced
Spending must be disciplined
Culture must be aligned
Product must stay ahead of the market

This supports projecting a net margin of 30% before 2030 and 40% by 2035.

My assumptions for the following metrics are:

• Share buyback impact: 1%

• P/E multiple: 25×

• Discount rate: 10% (my standard hurdle rate; allocation is driven by margin of safety and conviction)

• Model horizon: 10 years

Resulting intrinsic value:

$81.46 per share

I also ran a stress-test scenario with just 10% growth, well below market expectations, as a “floor” case.

Even then:

$28.58 per share

The conclusion is obvious:

$PATH Is a True Asymmetric Opportunity

Trading currently at only $19.04

The downside is limited.
The upside is massive.
The business is misunderstood, discounted, and entering a new capability wave just as enterprise AI adoption is beginning to inflect.

I’ve initiated a position and plan to add more over the coming months.

Remember to always do your own research. This write-up is an oversimplification; I didn’t really cover everything necessary to make a good investment decision, but I hope it serves as a good start. If you want to dive deeper, definitely check this post out: https://x.com/jakebrowatzke/status/1990253585649533373?s=46&t=n6nb9-cG0yPD_84M3dceIA and follow @jakebrowatzke

.

Sam Altman (CEO of OpenAI):

“Are we in a phase where investors as a whole are over-excited about AI? My opinion is yes.”

Jeff Bezos (Founder of Amazon):

At a tech event, he described the current AI surge as a “good kind of industrial bubble”:

“This is kind of an industrial bubble as opposed to financial bubbles … the ones that are industrial are not nearly as bad, they can even be good.”

Ed Zitron (Tech writer and critic):

Declares the 2020s AI boom a bubble and predicts it will pop.

Michael Cembalest (Analyst):

Describes today’s AI boom as a “vendor-financing circle” — capital raised by one set of companies is recycled through others, raising concerns of financial engineering rather than innovation.

A Bank of America Global Research survey found that 54% of fund managers believe AI stocks are in a bubble, versus 38% who do not.

Clearly, influential voices see bubble dynamics forming.

Is AI a Bubble?

Let’s break down the causes of fear, skepticism, and the foundation of this bear thesis.

I. What is a Bubble?

A bubble occurs when an asset, or an entire asset class, appreciates far beyond what is justified by its underlying economics. For businesses, this happens when expectations of future cash flows diverge widely from what the company can realistically deliver within a reasonable timeframe.

So, are some AI businesses in bubble territory, for sure they are. Some, just as always, won’t produce the necessary economics to justify their valuations or even their existence. Is this an issue?, absolutely not, this is the way that entrepreneurship and business reality works, and has worked for a while.

The real question isn’t whether some companies will fail, some will, but whether the entire AI sector is structurally overvalued, and whether the specific companies you own are driven by hype or by durable economics.

II. The Three Elements Behind the “AI Bubble” Narrative

1. Adoption & economic impact

2. Financing of the buildout (vendor financing and circular capital flows)

3. Economic feasibility (profitability/scalability of AI data centers, especially NeoClouds and GaaS)

1) Adoption & Economic Impact

The Dot-Com Analogy (and Why It Breaks Down)

Many arguing that “AI is a bubble” reason by analogy to the 2000 dot-com boom, saying “history rhymes.” But the analogy breaks once you examine what’s actually happening.

During the dot-com era, infrastructure came first and demand lagged by nearly a decade. Massive over-investment in fiber created “dark fiber,” and plenty of dot-com businesses never turned a profit. Adoption was slower than expected, use cases immature, and real economic gains modest.

The key difference in the AI revolution is actual demand, driven by faster adoption and clear economic benefit. AI is being deployed at an unprecedented pace, and in many cases it delivers immediate, measurable returns.

I don’t want to dismiss concerns about the technology itself; some are valid. There are legitimate questions about the ultimate capacity and scalability of AI models. If AI proves less transformational than expected, or if models hit a scaling wall and fail to reach the required capabilities, the industry could face real headwinds.

Adding to the debate, Andrej Karpathy recently discussed these issues in a podcast, questioning current model architectures, their economic impact, and the assumptions around progress toward AGI.

I won’t touch on every point here, but after watching the interview twice, I don’t think Karpathy is pessimistic about AI itself. He is skeptical of the idea that AGI will arrive tomorrow, replace humans, and instantly drive 20% GDP growth.

He highlights model limitations, the breakthroughs still needed for AI to become more broadly useful, and the fact that humans do not want or need AI to replace us; we want to work with it.

Karpathy’s view is that AGI may be a decade away, not a year.

Many took his comments as “the AI-bubble-bursting podcast,” much like the overreaction to DeepSeek’s efficiency breakthrough. In reality, his perspective is bullish: it reflects a field that is maturing while still driving rapid innovation. And it doesn’t change the fact that AI adoption keeps accelerating, with a growing share of the global economy already running on it.

Adoption and Economic Benefit Are Already Here

Adoption is broad, measurable, and economically meaningful. The largest AI spenders, including Meta, Microsoft, Amazon, and Google, are integrating AI into the core of their business models. Their combined CAPEX over the past twelve months exceeds about 291 billion dollars, against roughly 359 billion in combined net income. At that scale, there are no structural red flags.

We can also look at current use cases and see tangible, often immediate, economic rewards:

Consumer and enterprise software: search, ranking, recommendations; ad targeting; copilots improving output and reducing labor hours.

Commerce and logistics: Amazon’s recommendations, robotics, and forecasting; retailers’ dynamic pricing and shrink reduction; route optimization cutting fuel use.

Finance and insurance: fraud detection saving billions; JPMorgan’s COiN platform saving hundreds of thousands of legal hours; instant underwriting and claims.

Healthcare and life sciences: radiology-level models; accelerated drug discovery; documentation copilots freeing clinician time.

Energy and industrials: predictive maintenance; digital twins; grid forecasting for renewables.

Media and entertainment: generative tools reshaping creative workflows; procedural content in gaming.

Public sector and defense: cybersecurity, satellite analysis, decision support.

So yes, there are legitimate questions about the limits of current AI models, but there should be no doubt that today’s systems already generate real, tangible economic value for the companies that know how to deploy them effectively.

2) The “Circular Money Loop”

Some critics argue that Nvidia and other chipmakers are fueling a “circular money loop.” The claim is that Nvidia invests in AI startups such as OpenAI, xAI, and Anthropic, and those startups then use that capital to buy Nvidia GPUs, inflating valuations and repeating the cycle.

The key question is whether this is a closed loop or an open system. The main rebuttal is that real cash flows enter from outside. It is similar to an automaker providing financing to a dealership: if there are no customers, the money just circulates inside the loop. But once customers buy cars, external cash enters the loop with genuine demand. The same logic applies here. AI startups are not buying GPUs in a vacuum; they serve clients, run workloads, and generate revenue that flows back into the system.

Cross-investment between hardware, software, and cloud providers has long been part of platform industries. Apple and TSMC collaborate to secure advanced process nodes. Microsoft’s partnership with OpenAI guarantees Azure utilization. AWS co-funded startups that anchor workloads on its cloud. Even in traditional manufacturing, automakers and equipment makers often support downstream partners to accelerate distribution.

Nvidia’s ecosystem strategy follows the same pattern. By funding companies that build products and services around its GPUs, it strengthens the broader software stack, including CUDA and TensorRT, while seeding workloads and data-center capacity for future hardware. This approach echoes Intel’s past work on compiler optimization and AWS’s early support of AI startups that later became core customers. The goal is straightforward: accelerate real demand that would have emerged naturally.

Markets, as George Soros noted, are reflexive. Feedback loops exist, but not all are harmful. Healthy reflexivity amplifies fundamentals: better GPUs lead to better models, better products, more users, higher revenue, and more compute demand. Unhealthy reflexivity fuels speculation, where inflated valuations fund unprofitable ventures with little real usage.

So far, Nvidia’s ecosystem looks much closer to the healthy version. Its investments seem to amplify real adoption and infrastructure build-out rather than paper demand. Still, investors should recognize that reflexive growth can reverse if fundamentals weaken.

In short, what some call a “circular money loop” looks more like a strategic feedback system: capital reinforcing ecosystem growth. Nvidia’s role in AI’s expansion reflects a platform leader investing to strengthen and extend its technological base, not the financial engineering seen in earlier speculative cycles.

3) Economic Feasibility: Profitability and Scalability of NeoClouds and GaaS Models

Concerns about the economic feasibility of GPU-as-a-Service (GaaS) and NeoCloud data centers have grown louder, driven by a mix of misunderstanding of GPU economics, technology, flawed assumptions, and, in some cases, biased analysis.

The Critics and Their Claims

One of the most vocal critics has been legendary short-seller Jim Chanos, who argues that GaaS operators can’t earn sufficient returns on capital. When someone with his record raises questions, it’s worth listening, but his framework appears misapplied. He uses total CAPEX to calculate ROIC, rather than the portion of capital already producing cash flows. That’s like judging a hotel’s returns before it finishes construction.

Chanos also criticizes depreciation schedules, assuming GPU hardware will be obsolete within two or three years. That assumption ignores both historical precedent and the interplay between software and hardware improvements. I’ll return to that point shortly, but suffice to say: the evidence doesn’t support such an accelerated write-down.

Other critiques have come from The Information and The Financial Times, which questioned the profitability of GaaS models. The Information reported that Oracle lost $100 million on its AI cloud segment last quarter. That was likely a timing effect, a temporary mismatch between new capacity going live and revenue recognition as customers ramp up workloads. Oracle’s own commentary in its latest AI World conference effectively put those concerns to rest.

Meanwhile, the Financial Times argued that GPUs are “too expensive to buy and too cheap to rent.” Their math was flawed: they compared the price of a full 8-GPU node to the rental rate of a single GPU, an 8x error. Such oversights feed the perception of unsustainable economics where none exists.

The Obsolescence Misconception

Another persistent narrative is that GPUs will become obsolete every two or three years, as each new generation dramatically outperforms the last. Some suggest that future chips, like the Rubin CPX, will render current generations, such as GB300, uneconomic overnight.

If that were true, the entire NeoCloud sector would be at risk. But history points to a more measured reality: a regression to the mean.

Server hardware has historically remained productive for 7–9 years. Microsoft Azure’s own retirement schedule shows this clearly:

The K80, P100, and P40 GPUs (launched between 2014–2016) will retire in 2023, implying an 8–9 year life.

The V100 (launched in 2017) will retire in 2025, around 7.5 years later.

The A100(launched in 2020) and it remains economically viable.

That aligns closely with long-term IT depreciation norms. GPUs don’t vanish when the next model launches, they simply move down the value chain to workloads that don’t require the latest performance tier.

The Three Goals and the Three Forces Driving AI Economics

At its core, the AI industry is pursuing three fundamental goals:

A. Build smarter models.

B. Find and scale real-world use cases.

C. Drive down the cost per token.

These goals, in turn, activate three interacting forces that define the long-term economics of GPU infrastructure: demand growth, software optimization, and hardware innovation.

A. Demand Growth - The Expanding Workload Spectrum

The push to discover and scale new use cases continually broadens demand for compute. Every successful application (search, copilots, generative media, robotics) creates new workloads and revenue streams. As more industries integrate AI into operations, aggregate compute demand grows faster than supply expansion.

This persistent imbalance ensures that nearly all available hardware remains valuable. New GPUs power the frontier, massive model training, high-speed inference, while older chips cascade down to less intensive workloads where they still generate strong returns. Demand diversity stabilizes the system.

B. Software Optimization - Extending Hardware Lifetimes

The industry’s drive to reduce cost per token channels directly into software efficiency. Every layer of the stack (compilers, kernels, memory scheduling, quantization, fine-tuning algorithms) improves throughput without requiring new hardware.

Most of the gains in tokens-per-second efficiency today come from software, often by orders of magnitude compared to hardware alone. Continuous tuning by hardware vendors, hyperscalers, open-source contributors and AI labs, keep legacy GPUs productive far beyond initial expectations.

In effect, software innovation acts as economic gravity: it pulls down the effective cost of compute and extends the profitable lifespan of installed hardware.

C. Hardware Innovation - The Step Function of Capability

New architectures (like Nvidia’s Blackwell and Rubin) deliver large efficiency jumps but require long deployment and integration cycles. Power, cooling, and software adaptation slow immediate adoption. As a result, new hardware doesn’t instantly obsolete the old, it gradually shifts workloads upward along a capability curve.

Frontier training moves to the latest generation, while inference and specialized workloads continue profitably on prior ones. This interplay between innovation and inertia keeps the overall system balanced and capital-efficient.

The Interlocking System

These three forces and goals reinforce one another in a self-stabilizing loop:

The search for smarter models and new use cases expand demand.

The need for profitability drives software optimization.

Hardware innovation provides the next step in capability, which feeds back into new use cases and more demand.

Together, they create a layered equilibrium: demand outpaces supply, software stretches the useful life of capital, and hardware advances reset the frontier. The result is a sector that compounds through iteration rather than bubbles through speculation.

Fears about the economic viability of GaaS stem more from misunderstanding than reality. Demand is real and broad-based. Depreciation cycles are stable and supported by precedent. And cost-per-token reductions are coming primarily from software, not just hardware refreshes.

The result is an industry where nearly all compute remains economically useful, newer hardware drives the frontier, and older hardware serves the long tail. Far from fragile, the economics of GaaS look increasingly sustainable.

III. Conclusion

The question isn’t whether AI contains pockets of speculation, it does. The real question is whether the foundation beneath it is real, durable, and economically sound. Across all three dimensions: adoption, financing, and feasibility. The evidence points toward resilience.

Adoption and economic impact are already visible across every major sector. AI isn’t waiting for customers; it’s embedded in products, workflows, and infrastructure.

Financing of the buildout reflects strategic ecosystem development, not financial circularity. Nvidia, Microsoft, Amazon, and others are seeding complementary demand through partnerships and investments. Capital here is not being recycled in a vacuum; it’s reinforcing real economic activity driven by customers, workloads, and adoption.

GPU-as-a-Service operators and NeoClouds are showing early signs of sustainable unit economics. Depreciation cycles are likely to revert toward historical norms, while cost-per-token efficiency continues to improve, driven primarily by software innovation. The ecosystem’s self-balancing loop, rising demand, smarter software, and steady hardware progress, ensures that nearly every watt of compute finds productive use.

AI as an industry is not in a bubble. That doesn’t mean a bubble can’t form. It’s important to stay vigilant and know what to look for. It’s equally important to keep in mind the economic realities not only of the industry but of any individual company we choose to invest in.

Thanks for reading.

When a business is formed, if the founders are mission-driven, their vision is clear, their will is stronger than steel, and the core team is aligned, then sometimes a meta-core is born. The meta-core is the soul of the business: a fusion of mission, vision, alignment, and willpower. It is the non-physical force that drives the enterprise forward.

Given enough time and nurture, this soul begins to materialize into what can be called the company’s DNA. This DNA is not a rigid structure but a guiding pattern that enables the business to operate beyond what is normally possible. Over time, the DNA embeds itself into every level of operations and takes form as an operational algorithm—a repeatable way of doing things that improves with every cycle.

If consistency and execution persist, the operational algorithm develops into a powerful economic engine. That engine, through compounding, eventually produces what Warren Buffett famously called the moat—the durable advantage that protects great companies from competition.

The problem is that moats, once visible, are rarely mispriced. The investment edge lies not in chasing obvious moats, but in identifying the operational algorithm before it fully matures into great economics.

The final manifestation of the meta-core is strong financial outcomes: expanding margins, high returns on capital, operational leverage, free cash flow, etc. These are effects, not causes, the fruit of a strong economic engine.

Therefore, the task of the thoughtful investor is not only to measure outcomes but to search for the meta-core itself.

Read the transcripts. Study the 10-Ks. Watch the interviews. Ask the right questions.

Reality cannot be directly perceived; it can only be interpreted. Our interpretations depend on the mental models we use, and those models can be refined, sharpened, and improved.

Search for the truth hidden beneath the obvious.

The long-term vision of the core team is more important than next quarter’s revenue growth.

The consistency and integrity of what management says and does is more important than short-term margins.

The addition of key hires is more valuable than short-term cash flows.

The operational algorithm is more foundational than the moat.

Value creation is more important than dilution.

Long-term business trends matter. Execution is everything.

And buying below a conservative estimate of intrinsic value remains essential—because even the greatest businesses, with fine-tuned operational algorithms, will eventually stumble in the eyes of the market, creating rare opportunities for those who have studied the soul of the enterprise.

$IREN is one of the most asymmetric plays in the AI arena, with mid-term upside potential of 10–20x.

I generally avoid Bitcoin miners. Most are just levered plays on BTC—bloated with debt, addicted to dilutive capital raises, and built on structurally weak models. They chase hype and live or die by Bitcoin’s mood swings.

But Iris Energy ($IREN) is different.I reiterate ,IREN might be one of the most asymmetric setups in the public markets today.

You’re getting a fully built, world-class infrastructure platform—one that could generate $500M+ in EBITDA from Bitcoin mining alone—while paying almost nothing for the embedded AI colocation optionality.

At today’s valuation, I think this is one of those rare 5–10x setups where the downside is unusually capped, and the upside? Difficult to model.

🏗️ Business Overview

Iris Energy began as a high-efficiency, renewables-powered Bitcoin miner—but that’s no longer the full story.

They’re now pivoting hard into AI and HPC colocation, and they already own the power, land, and infrastructure to support hyperscalers at scale.

• Legacy business: Clean BTC mining (100% renewables), with a projected $500M+ run-rate EBITDA in 2025

• Next chapter: Over 3 GW of fully owned infrastructure—including BTC mining sites—with 2.4 GW expected to be energized in the next 12–18 months

⚡ DC Sites Portfolio

Childress (750 MW)

• Phase 5 (150 MW) energization imminent; substation transformer installed

• Horizon 1 on track for Q4 2025 delivery

• Second transformer on site enables full 750 MW energization and redundancy

Sweetwater 1 (1.4 GW)

• Site work and substation construction underway

• Energization: April 2026

Sweetwater 2 (600 MW)

• Design completed for direct fiber loop connecting to Sweetwater 1 (2 GW hub)

• Energization: Late 2027

They’re not speculating on demand—they’re actively engaged with AI players. Early signs point to meaningful traction.

📈 Unit Economics (Colocation)

This is where the leverage gets interesting:

• Annual revenue per MW: $1M–$1.5M

• Gross margins: 65%–75% (depending on deal structure)

• Contract length: 10–15 years

• Customer base: Hyperscalers, LLM startups, AI training clusters—non-cyclical, sticky demand

This isn’t a miner anymore—it’s a digital landlord at the center of the AI compute land grab. And they already own the land.

💰 Financial Position

• No toxic debt

• Fully self-mined (no hosting risk)

• Cash flow–positive core business

• Full control of 3+ GW of sites with substations and power rights secured

IREN is capital-disciplined. They didn’t chase the AI hype—they just happen to already own the infrastructure everyone now needs.

👥 Management

Founder-led, with deep experience across infrastructure, power, and capital markets.

They’ve weathered multiple crypto winters without diluting shareholders into oblivion.

The AI shift isn’t a desperation pivot—it’s a logical evolution based on hard assets they’ve quietly built over years.

Execution has been sharp: low SBC, no hype cycles—just build, energize, monetiz

e.

🧾 Final Thoughts

This isn’t some microcap flyer with a whitepaper and a dream.

IREN is producing real EBITDA, owns 3 GW+ of high-quality land and power, and is now onboarding into the highest-margin compute supercycle of our generation.

It’s a sleeper AI infrastructure play disguised as a Bitcoin miner. Most haven’t noticed yet.

🧮 Valuation Math

Peer comps in AI infrastructure trade north of 11× P/Sales.

If we assume:

• $1.5M revenue per MW

• 3,000 MW portfolio

  → That’s $4.5B in annual revenue once fully energized and leased.

Assigning a 10× sales multiple puts the potential market cap at $45B.

That’s 19x upside from today’s valuation or $9.8 p/ share today to $186 in 24-30 Months.

Lemonade is a multi-line insurtech company offering a broad range of property, casualty, and life insurance products. Founded in 2015 and based in New York, NY, LMND uses artificial intelligence (AI) and behavioral economics to streamline the insurance process, targeting a digital-first, younger demographic with a mission to “rebuild insurance from the ground up.”

Core Products:

Renters, homeowners, auto, pet, life, and car insurance, available primarily in the US with a small presence in Europe (e.g., Germany, France, Netherlands). Target Market: Tech-savvy consumers, predominantly under 35, seeking simple, affordable insurance through a digital platform, with over 2.43 million customers as of Q3 2024.

Technology Utilization:

AI Maya: An LLM that handles policy issuance across all product lines, enabling instant coverage.

AI Jim: An LLM that processes claims in as little as two seconds, automating payouts and reducing human intervention.

Proprietary platform “Blender” integrates all operations, using real-time behavioral data (e.g., language, response times), geospatial analysis, and weather data for precise underwriting.

Revenue Model:

Premiums from policyholders, with a significant portion (around 55%) ceded to reinsurers to de-risk operations. A “Giveback” model donates unclaimed premiums to charities chosen by customers, enhancing brand loyalty but not directly contributing to revenue. Key Differentiators: Broad product portfolio allows cross-selling opportunities (e.g., 17% of pet insurance sold to existing renters customers). Heavy reliance on AI for both underwriting and claims, aiming for low operational costs (over 2,000 customers per employee vs. peers’ 250-450). Digital-only, no-agent model contrasts with traditional insurers’ reliance on intermediaries.

Risk Management:

Cedes a higher-than-average 55% of premiums to reinsurers, reducing volatility and capital reserve needs, with plans to lower this as AI models improve and stabilize cash flows (e.g., through life insurance expansion).

2. Oscar Health (OSCR) Business Model Overview:

Oscar Health is a technology-driven health insurance company focused on providing individual and family health plans, primarily through the Affordable Care Act (ACA) marketplace. Founded in 2012 and headquartered in New York, NY, Oscar aims to simplify the healthcare experience for consumers by leveraging data and technology to improve member engagement and reduce costs.

Core Products:

Individual and family health insurance plans (ACA marketplace). Small group plans and Medicare Advantage in select markets.

Target Market:

Individuals and families seeking affordable health insurance, often through government-subsidized marketplaces, with approximately 1.7 million members across 20 states as of December 31, 2024. Technology Utilization: Uses a proprietary platform to enhance member experience with features like telemedicine, personalized care navigation, and a user-friendly app for managing claims and finding providers. Data analytics to optimize care delivery, focusing on high-engagement touchpoints (e.g., 24/7 doctor access via app). Emphasis on transparency and simplicity, contrasting with the complexity of traditional health insurers.

Revenue Model:

Premiums from policyholders, supplemented by government subsidies under the ACA. Aims to reduce medical loss ratios (MLR) through preventative care and technology-driven efficiency, though still reliant on reinsurance to manage risk.

Key Differentiators:

Focuses exclusively on health insurance, a highly regulated and complex market. Prioritizes member experience over pure cost-cutting, aiming to retain customers in a competitive ACA landscape. Less emphasis on underwriting automation compared to property and casualty insurers like LMND, ROOT, and HIPPO; more focus on care coordination.

Risk Management:

Relies on reinsurance to offset high medical claims, typical in health insurance, and operates within ACA risk adjustment programs to balance high-cost patients.

3. Root Insurance (ROOT) Business Model Overview:

Root Insurance is a tech-driven auto insurance company that uses telematics to assess driver risk and price policies based on individual behavior rather than demographics. Founded in 2015 and based in Columbus, OH, ROOT operates in all 50 states, focusing on disrupting the auto insurance market with a data-centric approach.

Core Products:

Auto insurance (personal lines only, no commercial or multi-line offerings).

Target Market:

Drivers willing to share driving data via a mobile app, particularly those who believe they are low-risk and can benefit from behavior-based pricing, with a growing customer base in a $380 billion US auto market.

Technology Utilization:

Telematics via a mobile app tracks driving habits (e.g., braking, speed, time of day) during a test period to set premiums, avoiding traditional factors like age or gender. Machine learning refines risk models, aiming for precision in identifying low-risk drivers. Digital-first platform eliminates agents, focusing on app-based policy management and claims.

Revenue Model:

Premiums from policyholders, with a focus on retaining low-risk drivers to improve loss ratios. Minimal reliance on reinsurance compared to LMND, as telematics aims to inherently reduce risk exposure by targeting safer drivers.

Key Differentiators:

Niche focus on auto insurance with telematics as the core differentiator, unlike LMND’s multi-line approach. Pricing based on actual driving behavior rather than proxies, appealing to safe drivers seeking lower rates. No physical infrastructure or agent network, fully app-based.

Risk Management:

Relies on telematics to self-select low-risk drivers, reducing claims frequency and severity without heavy reinsurance dependence, though this exposes ROOT to market cyclicality and adoption risks.

4. Hippo Holdings (HIPPO) Business Model Overview:

Hippo Holdings is a tech-enabled home insurance company focused on proactive protection rather than just reactive claims handling. Founded in 2015 and headquartered in Palo Alto, CA, HIPPO uses technology to mitigate risks and enhance customer experience in the home insurance market, with some expansion into adjacent lines.

Core Products:

Homeowners insurance (primary focus), with ancillary offerings like auto, pet, and small commercial insurance in select markets.

Target Market:

Homeowners seeking modern, tech-driven insurance with added value (e.g., home maintenance services), targeting a growing segment of the $120 billion US home insurance market.

Technology Utilization:

Smart home technology (e.g., sensors for leak detection, fire prevention) offered free to policyholders to reduce claims, paired with data from satellites and public records for underwriting. Digital platform streamlines policy issuance and claims, though less AI-reliant than LMND or ROOT. Partnerships with home service providers (e.g., maintenance, inspections) enhance value proposition.

Revenue Model:

Premiums from homeowners, with additional revenue potential from partnerships and service fees (e.g., Hippo Home Care). Moderate reinsurance use to manage catastrophe risk, common in home insurance due to natural disasters.

Key Differentiators:

Proactive risk mitigation through smart home tech sets it apart from traditional home insurers and LMND’s broader focus. Smaller product range than LMND, but more specialized than ROOT’s auto-only model. Combines insurance with home protection services, aiming to reduce loss ratios and attract customers.

Risk Management:

Uses reinsurance to offset catastrophe risks (e.g., wildfires, hurricanes), while smart home tech aims to lower claims frequency, balancing reactive and proactive strategies.

Comparative Analysis

Scope and Diversification:

LMND’s multi-line approach offers the most diversification, reducing reliance on a single market compared to ROOT (auto), HIPPO (home), and OSCR (health). This allows LMND to cross-sell and tap multiple revenue streams, a flexibility others lack.

Technology Application:

LMND and ROOT are the most AI- and data-intensive, but LMND applies it across products, while ROOT focuses solely on telematics for auto. OSCR uses tech for engagement rather than underwriting, and HIPPO blends tech with physical devices, a hybrid approach.

Risk Management Strategy:

LMND’s heavy reinsurance (55%) contrasts with ROOT’s reliance on telematics to inherently lower risk, HIPPO’s mix of reinsurance and prevention, and OSCR’s use of ACA mechanisms alongside reinsurance. LMND’s approach prioritizes growth over immediate profitability, while others balance risk differently.

Customer Experience:

OSCR emphasizes healthcare navigation, LMND focuses on speed and simplicity, ROOT targets fairness in pricing, and HIPPO adds value through home protection. Each caters to distinct customer needs within their niches.

Scalability:

LMND’s broad platform is the most scalable due to its multi-line potential. ROOT and HIPPO are constrained by their niches, while OSCR’s scalability is tied to ACA market dynamics and regulatory changes.

Conclusion

The business model differences reflect each company’s strategic focus:

OSCR is a health insurance play with a consumer-centric, regulated model.

LMND is a versatile, AI-driven multi-line insurer with broad growth potential.

ROOT is a specialized auto insurer betting on telematics for precision. HIPPO is a niche home insurer with a proactive, tech-enhanced approach.

In my opinion, LMND’s diversified model offers the most balanced opportunity and provides the greatest long-term growth potential.

Subscribe now