Why AI Agents Can Beat the Incumbents
Leo, an AI-native startup, is building multi-agent systems to automate end-to-end procurement processes for enterprises by coordinating work across multiple departments, stakeholders, and external systems that traditional procurement software cannot handle. The company differentiates itself from incumbents by owning the entire job workflow rather than being confined to a single system of record, and builds customer trust through human-in-the-loop approaches before scaling to fully autonomous negotiations.
Summary
The podcast discusses why AI-native startups can compete against established software incumbents in the enterprise space, using procurement as a detailed case study. Elena Berger interviews Seema Amble from A16Z and Vlad Mirovskii, CEO of Leo, exploring how AI agents can beat legacy vendors like Salesforce and SAP.
The core problem identified is that 80% of procurement work happens outside systems of record—in emails, spreadsheets, contracts, and conversations between suppliers and internal stakeholders across legal, finance, engineering, and procurement departments. When a Boeing engineer needs a bolt, the process involves inventory checks, sourcing, RFQ drafting, quote evaluation, negotiation, order confirmation, shipment tracking, and invoice processing, often coordinated across hundreds of stakeholders and involving critical decisions worth hundreds of millions of dollars if delays occur.
Seema outlines four levels of AI agents: retrieval agents (pulling information), process agents (applying defined rules), policy agents (applying judgment within guidelines), and principal agents (making discretionary decisions). Incumbents typically offer only retrieval and basic process capabilities because of internal conflicts—automating work end-to-end cannibalizes their workflow software sales, creating competing product teams and sales challenges.
Leo's advantage lies in owning the entire end-to-end arc rather than being confined to one system. The company uses a human-in-the-loop approach where agents start with lower-risk negotiations (under $50K) and gradually earn trust for larger deals through continuous feedback. For multi-million-dollar negotiations, human experts remain involved in real-time, with agents providing preparation analysis and real-time market insights during negotiations.
The company operates as a multi-agent system where different specialized agents handle inventory checks, sourcing decisions, RFQ generation, response analysis, negotiation strategy, and contract management—all coordinating to complete complex procurement tasks autonomously or with human oversight depending on risk and complexity levels.
Vlad explains that while engineers could build basic quote-to-system automation in eight hours, achieving 100% performance rather than 70% requires integration expertise, memory management, workflow orchestration, and vertical domain knowledge that enterprises struggle to maintain internally. The transcript includes Leo's bootstrapping strategy of always staying ahead of customer needs, starting with document retrieval three years ago and continuously advancing to more complex autonomous capabilities.
The discussion also explores the future where both buyers and suppliers use agents on the same platform, creating coordinated automation on both sides of transactions. Importantly, the parties' incentives align on 5,000+ operational tasks (speed, accuracy, efficiency) even if they disagree on price—enabling collaborative agent-to-agent work. The transcript concludes with Vlad's perspective on why procurement—emotional, boring, and economically massive—represents a trillion-dollar opportunity, illustrated through Leo's Buyers and Sellers summit that attracted over 100 C-level procurement leaders.
About this episode
a16z’s Seema Amble and Elena Burger sit down with Lio co-founder and CEO Vladimir Keil to ask where AI-native startups have an advantage when incumbent software companies already own the customer, the data, and the system of record. Their answer comes down to the work that happens outside those systems. In procurement, a final price in an ERP can hide hundreds of emails, spreadsheets, supplier conversations, engineering analyses, and decisions across legal, finance, and operations. Vlad explains how Lio uses multi-agent systems to take on more of that end-to-end work, from sourcing and RFQs to negotiation, shipment tracking, and invoices. They also discuss how enterprises learn to trust agents with increasingly consequential decisions, why the last 20% of an internal AI build can require most of the effort, and what happens when both buyers and suppliers have agents working on their behalf.
Key Insights
- Vlad argues that 80% of procurement work occurs outside enterprise systems of record in emails, spreadsheets, and conversations, meaning incumbents who layer AI only onto their existing software cannot capture the complete job workflow.
- Seema identifies that incumbent software vendors face internal conflicts between automating work end-to-end versus maintaining workflow software sales, causing different product teams to compete and preventing them from advancing beyond retrieval and basic process agents.
- Vlad claims that while engineers can build basic document automation in eight hours, achieving full production-quality automation requires the remaining 20% to involve integration, orchestration, and vertical domain knowledge—making the last 20% consume 80% of actual effort.
- Leo uses a trust-building strategy where agents start with low-risk, high-volume negotiations (under $50K) that companies previously ignored due to capacity constraints, then progressively earn trust for larger deals through continuous feedback loops.
- Vlad explains that in complex direct procurement negotiations worth millions of dollars, agents primarily do 90% back-office preparation work (analyzing drawings, pricing, supplier reliability, market conditions) while humans handle real-time negotiation, with agents providing real-time market insights during discussions.
- Vlad argues that buyers and suppliers have aligned incentives on 5,000 operational tasks (speed, quality, efficiency) even when they disagree on price, enabling both sides to deploy agents that automate coordination rather than creating adversarial dynamics.
- Seema notes that the durability of vertical AI companies comes from locking in customers through dependency and performing more of the overall work, not from predicting specific technical moats, with stickiness and value creation preceding defensibility.
- Vlad claims procurement represents a trillion-dollar opportunity because it combines three elements: high emotional frustration among practitioners, boring/stagnant technology (no major changes in 20+ years), and massive business impact where 1% procurement savings equals 10% revenue growth in P&L impact.
Topics
Transcript
If you want to build an aircraft, you need to procure thousands of suppliers. Someone sends a confirmation of like, hey, sorry, like this part is going to arrive two weeks later. And if they missed this email, hundreds of millions of them. Procurement historically may have been more in a box. And now it's like, okay, it's touching legal, it's touching finance, it's touching a bunch of different software systems and people. The opportunity for the AI-native startup is to say, we're going to own that entire end-to-end arc. No company and no enterprise starts with fully autonomous negotiation agents from day one. Why? Because they don't trust us and they don't trust the technology from day one. And…
Full transcript available for MurmurCast members
Sign Up to AccessMore from The a16z Show
The Top 100 Consumer AI Apps: Who’s Actually Paying?
A16Z's seventh edition Top 100 Consumer AI Apps Report reveals that while half of Americans use AI, only 4.5% pay for subscriptions, with spending heavily concentrated among developers and power users. The report highlights emerging personal agents as a major trend, significant divergence between ChatGPT, Claude, and Gemini, and substantial untapped opportunities in categories like dating, social, shopping, and entertainment.
David George & Jack Altman on AI, Autonomy, and the Next $25 Trillion
David George from Andreessen Horowitz argues that AI, autonomy, and robotics will create over $25 trillion in market value over the next decade, with success across multiple layers of the stack—frontier models, open source, and applications will all flourish due to massive supply constraints and early-stage demand penetration among knowledge workers.
Beyond the God Model | Alex Atallah & Amjad Masad
Alex Atallah (OpenRouter) and Amjad Masad (Replit) discuss how the future of AI development lies in specialized, diverse models working together rather than relying on single AGI-like models. They argue that enterprises need independence from foundation model providers through model diversity, cost efficiency, and ownership of their own AI capabilities, while also addressing the risks of general-purpose agents and the importance of model specialization.
The $1 Trillion AI Buildout | State of Markets
A16Z partners discuss the $1 trillion AI infrastructure buildout, arguing it's driven by real earnings growth rather than inflated valuations, with adoption still extremely early at the enterprise level. They highlight opportunities across consumer agents, robotics, autonomous vehicles, and enterprise diffusion, while noting that the market's 90% gain since ChatGPT reflects fundamental business performance rather than speculative excess.
The Personal Agent Race Is Here | Anish Acharya & David Pawlan
A16Z's Anish Acharya and Assistant Benchmark creator David Pawlan discuss the explosive growth of personal AI agents, exploring their capabilities across email, travel, and finance use cases, the infrastructure needed for agent-to-agent interactions, and how these systems will reshape commerce and consumer behavior.