Turn Your Agreements Into Intelligent Data with Docusign Iris

Our mission is to explore how artificial intelligence is fundamentally transforming the way organizations handle agreements. We want to show you how static dead documents are basically being turned into active intelligent assets using the Docuine AM platform — which is such a fascinating evolution to unpack because you know for the last couple of decades we really thought we were innovating by just digitizing paperwork. — Right. Like scanning a piece of paper. — Exactly. Taking a physical piece of paper and turning it into a PDF on a screen. But a PDF is still just a static picture of a document. It does not actually do anything. — Yeah. It's just pixels. Right. So what we are exploring today is intelligent agreement management. It is a total paradigm shift where the agreement itself becomes a dynamic functioning piece of your business operations. — Okay, let's unpack this cuz I want to get straight to the mechanism driving this. — It is so easy to just throw around the term AI as like a magic wand. But under the hood of the docyign platform is an

engine called docuine iris. Right. — Yes. Docuign Iris. And from my understanding, it is not just one massive chatbot that you blindly feed your contracts into. — Oh, far from it. Docycine Iris is the foundational agreement AI engine powering the entire Docyign AM platform and it functions using a hybrid approach. It actually combines two very different types of artificial intelligence to handle the complex reality of legal language. — Okay, break that hybrid approach down for me. What are the two sides doing? — So, on one side you have traditional AI. Traditional AI is highly deterministic. It is incredibly good at pattern recognition and extracting highly specific data. Think of it as a blood hound. — A blood hound. Okay, I like that. — Yeah. If you need to find a specific effective date or say extract a liability cap buried on page 47 of a master services agreement, traditional AI sniffs it out with nearperfect accuracy. It isolates the facts. — Okay, so the traditional AI is the finder. It is the extractor, — right? But finding a clause is very different from understanding its broader implications. And that is where the other side of the hybrid model comes in generative AI. — Ah okay. The generative side. — Exactly. Generative AI is probabilistic. It understands context, tone, semantic meaning. So while traditional AI finds the dense legal clause, the generative AI component synthesizes that information and creates entirely new content based on it. — Like what kind of content? like drafting a plain language summary of that clause so a non-awyer can actually understand it. — So it's almost like having a worldclass lightning fast parallegal who has read every contract ever written and instantly knows what is standard and what is risky. — That is exactly what it is like. Yeah, — that makes a lot of sense. You need the precision of the traditional AI so you we're not hallucinating dates or numbers, but you need the generative AI to make the output human readable. — Precisely. — But uh here's the elephant in the room. Contracts are arguably the most sensitive confidential documents an organization has. If Docyign Iris is this incredibly smart engine, how did it get so smart? Because you cannot just scrape the public internet to learn complex corporate law. — You absolutely cannot. You cannot rely on generic internet text to understand the nuances of a commercial real estate lease or a highstakes merger. — Right? — So, Doc Saniris has been built and trained on hundreds of millions of private consented agreements. It understands the actual practical language of global business because it has been genuinely trained on it. — But wait, the moment you say trained on private agreements, I guarantee every chief information security officer listening to us right now just started to panic. — Oh, for sure. — Right. If a company puts their highly confidential trade secrets into the Docuine AM platform, are they accidentally training an AI model that their competitors could somehow benefit from? That is the most critical question in enterprise technology today and the mechanism protecting that data is strict anonymization. Before any data is ever used to inform an AR model, the system actively strips out personal identifiers. — But what exactly gets stripped out? — We are talking about names, phone numbers, email addresses, social security numbers, bank details. It is all systematically scrubbed. Okay, so the AI learns the structure of a great indemnification clause, but it has no idea who actually signed it or what the specific financial figures were — precisely. And beyond that, third party AI models do not get to train on customer data at all. The data you put into the Docyign AM platform stays protected within that ecosystem. You get the benefit of a massive business trained AI without ever exposing your proprietary data to the public domain. Well, having a highly secure business fluent AI engine is an incredible foundation, but a foundation only matters if it changes how we actually work on a Tuesday afternoon. You know, — very true.

— So, let's look at how docysine Iris fundamentally alters the very first step of the agreement life cycle, the creation process. Because right now, preparing a document for signature is an intensely manual drag and drop nightmare. — It is entirely tedious. You upload a document, you manually scroll through to figure out where the signature lines are. You drag the date field over, you drag the initial boxes over. It takes forever. — Yeah, it does. — But Doc Sign Iris automates this completely through a mechanism called autotagging. — How does it know where things go? Is it just like looking for visual blank spaces? — No, it is parsing the text and the document structure simultaneously. When you upload a document, the AI automatically reads it and detects the agreement type. It knows immediately if it is a non-disclosure agreement versus a sales order. — Oh wow. — Yeah. And then it looks for specific text markers like the words effective date in brackets. It mathematically maps the coordinates of that text on the page and snaps the correct interactive digital field right onto that exact spot instantly. — That is huge. That completely eliminates the manual setup for standard contracts. — Yeah. But the creation phase also encompasses what I honestly consider the absolute worst customer experience in the world. The PDF marathon. — Oh, Yes, we have all experienced this, right? You were trying to open a bank account or apply for a permit and the organization forces you to download a static PDF. You have to print it out, fill it in by hand, realize you made a mistake, print it again, scan it, and email it back. It is maddening. — And it is equally maddening for the organization receiving it because what do they get? — Right. A messy scan. Exactly. They get an unarchable image file with messy handwriting which someone on their team now has to manually retype into their database which just introduces human error at every single keystroke. — So how does AI fix the PDF marathon — through AI assisted web forms? Docyign Iris essentially uses computer vision and structural analysis to look at that static clunky PDF. It understands the layout. So it sees a label that says address and a blank line next to it. — Okay. And it translates that visual space into dynamic HTML code. In a matter of minutes, it transforms a dead PDF into a guided mobile responsive web form. — Wait, without needing a team of software engineers to code the web page from scratch? — Exactly. Zero coding required. — Okay, so data collection is totally solved. But what about the actual heavy duty legal drafting? When two massive companies are negotiating, they are not filling out a web form. They are battling it out inside Microsoft Word. — They are. And that is why AI assisted review is brought directly inside Microsoft Word. You do not have to leave your word processor. As you are drafting or reviewing a thirdparty agreement, the AI acts as an active assistant right there in your document pane. — Wait, I have to stop you there. Are you telling me an AI is going to redline a multi-million dollar liability clause? Because every corporate lawyer listening just broke into a cold sweat. If the AI is hallucinating or just rewriting clauses based on what it thinks sounds good, an organization is going to lose total control of its legal risk profile. — Oh, if the AI were acting on its own, that would be completely terrifying. But that is not how AI assisted review operates. It is a highly constrained system. It strictly references your organization's pre-built playbooks and approved clause library. — Ah, okay. So, it is anchored to the company's established rules. — Exactly. It uses semantic similarity matching. It reads the draft agreement that the other party sent you. And it compares the meaning of their clauses against the meaning of your standard terms. — Oh, that's clever, — right? If it detects a deviation, say they want net 60 payment terms and your standard is net 30, it flags the risk. It then automatically suggests the specific pre-approved red line from your company's playbook to bring the language back into alignment. — So, it is not inventing law. It is just surfacing the exact language the general counsel already approved right when the reviewer needs it. — Precisely. It can also answer plain language questions about the document. This does not replace the lawyer at all. It acts as a highly diligent second set of eyes doing the tedious comparison work. It frees up legal professionals from spending hours doing manual lineby-line comparisons so they can focus their highly trained expertise on the strategic negotiations that actually require human judgment. Right? Which makes total sense. So the drafting is done, the language is locked, the red lines are accepted, and the agreement is perfectly prepared. But all of that precision is completely useless if the person receiving the contract just blindly signs it without understanding a word of it. Which brings us to the next

phase, right? The commit phase. — The commit phase. Yes. And the psychology here is truly fascinating because the data reveals a massive vulnerability in how we do business. Nearly 60% of people admit they have agreed to terms they did not fully understand. — 60%. And honestly, who can blame them? When you are presented with a 40-page commercial lease written in dense legal ease, and you are trying to run a business, you do not have the time or the specialized training to decode it. You just cross your fingers and sign, — which is terrible for trust and terrible for compliance. Interestingly though, only about 2% of people actually want to skip or skim these agreements. — Really, only 2%. — Yeah. The vast majority want to comprehend what they are committing to, but they are simply overwhelmed by the complex syntactic structures of the text. To solve this, docuign Iris uses its generative AI capabilities to provide an AI assisted summary. — Okay, so what does that look like for the signer? — So before you even get to the signature block, the AI synthesizes the entire document and generates a concise plain English summary of the key terms. It strips away the jargon and highlights the crucial obligations like how much is being paid, what the core deliverables are, and the timeline. — I can absolutely see how that drastically reduces friction. If I am the signer, my anxiety drops. I am not blindly clicking accept anymore. I actually grasp the essential information, allowing me to move forward and commit with real confidence. — Exactly. But confidence has to go both ways in a transaction. The person sending the contract needs absolute confidence that the right person is actually sitting behind the keyboard. And with the rise of AI generated impersonation, deep fakes, and synthetic media, establishing digital trust is harder than ever before. — This is a huge issue. I mean, if I can use software to generate a photorealistic video of anyone saying anything. How do you verify identity online without forcing someone to physically walk into a notary's office? — You leverage identity verification via livveness detection. When a high-v value agreement requires strict verification, the signer takes a live video selfie using their phone or webcam. — And how is that secure against a deep fake? — Could a fraudster just like hold up an iPad playing a deep fake video of the target to the webcam? — No, because livveness detection is not just taking a flat picture. It is measuring biological reality. The AI biometric technology maps the facial geometry in three dimensions. — Oh wow. — Yeah. It analyzes micro movements, depth perception, and how light reflects off the skin. A deep fake on a screen lacks those subtle subddermal depth cues. The system verifies that a live breathing human is physically present right then and there. And then it compares that biometric mesh against the photo on their governmentissued ID. — So it definitively links the physical human to the digital identity credential, basically scoop proofing the system. And from a user experience standpoint, they are just looking at their phone for 3 seconds. — Exactly. And the Docyign AM platform is even moving toward accepting mobile driver's licenses like the IDs you can securely store in your Apple wallet. You just tap your phone, pass the biometric livveness check, and the identity is proven. — That is amazing. It provides massively higher identity asurances without sacrificing speed or user convenience. — Exactly. — Now, here is where it gets really interesting. Most organizations think the process ends the moment that highly secured digital signature is captured. The champagne pops, the revenue is

booked, and everyone moves on. But a signed contract isn't the end of anything. It is the literal beginning of a business relationship with ongoing obligations that have to be managed. — And yet, management is where contracts usually go to die. — Totally. They get signed, downloaded as PDF, and then they disappear into the black hole of a shared drive, an obscure SharePoint folder, or an email inbox. — Yeah, I actually read a metric recently that it currently takes contracting professionals an average of 45 minutes just to find one specific agreement. If you are a procurement manager trying to audit a 100 vendor contracts, you are losing weeks of your life just using a search bar. — Because a PDF in a shared drive is unstructured data. to a computer it is just a dumb image of text. It has no searchable properties other than the file name. Right? — So to solve this the docuign AM platform features the agreement manager. It acts as a centralized intelligent repository for all your executed agreements. But it does not just store them like a digital filing cabinet. Docyign Iris scans every single document and uses entity extraction to turn that flat unorganized text into structured actionable data. — Break down entity extraction for me. How does it make text actionable? — Well, the AI reads the document and identifies the core components as distinct data points. It finds the governing law, the payment terms, the renewal clauses, and the specific liability limits. It essentially turns a massive text document into a highly organized database. — Okay, that is super powerful. — It is. Once that data is structured, you do not have to open 50 PDFs to find an answer. You can simply ask plain language questions across your entire repository of thousands of documents. — So, give me a concrete example of how a specific department uses this structured data in the real world. — Let's look at human resources. HR deals with a massive volume of independent contractor and gig worker agreements, often with varying terms and end dates. Using custom extractions, HR can instantly pull a report showing every gig worker engaged across the entire company exactly what their specific terms are and when their contracts expire all without opening a single file. — That is massive for compliance. But the financial impact seems most obvious in procurement. Procurement teams are managing millions of dollars in vendor spend. — Oh, procurement is where this technology pays for itself almost immediately. Yeah, — we have all experienced the pain of being locked into a bad software or supplier deal simply because someone missed a 90-day cancellation window hidden on page 12 of a contract. — Oh, absolutely. The classic software license trap. You buy enterprise software for your whole team. You sign a three-year deal. The champion who bought it leaves the company and everyone forgets about the auto renewal. — Exactly. So, let's say you have an autorenewing contract for 500 software lace notas. The agreement manager has extracted those terms and knows exactly when the cancellation window opens. But because the Docyign AM platform integrates with your other business systems, it can actually cross reference data. — Wait, how does that work? — It might look at your single signon provider and discover that your team is only actively using 300 of those 500 seats. — Whoa. So, the AI surfaces the discrepancy. It flags this massive cost-saving opportunity, warning you weeks in advance. Hey, you are about to autorenew for 200 empty seats. You need to rightsize this contract right now. — Yes, that level of visibility is unprecedented. — It is incredible. But it leads me to the ultimate question. The AI gives me this brilliant insight. It tells me I have 200 unused licenses and the deadline is approaching. Do I still have to manually draft an email to the vendor, manually negotiate a new agreement, manually route it for approval, and then manually update our spend management software? If so, the bottleneck is still kind of there. And you just highlighted the exact threshold between where AI has been and where it is going. You are describing the leap from an AI that functions as an oracle, giving you insight to agent actually taking autonomous action.

And that brings us to the cutting edge of the Docuine AM platform, Docusine AI agents. Okay, let's go deep on this because moving from an AI that just talks to you to an AI that executes multi-step workflows across different software platforms, that is a fundamental evolution in how work gets done. How do these agents operate? — Think of it as interacting with an incredibly capable digital assistant that understands your full historical agreement context and has the security permissions to execute tasks. Inside the Docyign AM platform, you do not just ask a question. You trigger a specific agent to run a complex playbook. — What kind of playbooks are we talking about here? — Well, there are several readyto- deploy agents that handle the heaviest administrative lifting. First, there is intake and triage. — Okay. — In a traditional legal department, highly paid lawyers spend hours every day just reading incoming requests to figure out who should handle them. The intake agent automatically reads incoming contracts, assesses the risk profile based on your parameters, flags exceptional terms, and automatically routes it to the exact right specialist or approver. No human traffic cop required. — That eliminates the inbox bottleneck immediately. — Exactly. Then you have smart redlinining. We talked earlier about AI suggesting changes in Microsoft Word. The smart redlinining agent takes it a massive step further by doing the actual markup automatically based on your past accepted terms. Wait. Automatically? — Yes. If the agent knows your company always pushes back on a specific indemnification structure because it has access to your entire contract history, it intercepts the new agreement and actively does the red lining for you, presenting you with a finished draft to approve. — It learns your negotiation style. That is wild. — It really is. And perhaps the most strategic one is relationship intelligence. — How does that one change the workflow? Imagine you are a procurement officer walking into a quarterly business review with a major long-term vendor. Traditionally, you would spend days hunting down the original master agreement, the five subsequent amendments, the pricing addendums, just trying to piece together the history. — Yeah. You're usually walking in blind and just hoping you remember the details. — Exactly. Instead, you trigger the relationship intelligence agent. It instantly accesses your entire repository, reads all the past signed agreements, synthesizes the deal history, and assembles a comprehensive plain English brief. You walk into that negotiation perfectly prepared with total leverage because the AI synthesized years of complex history in seconds. — But the true magic of an agentic system is that it does not just live inside one application, right? like an agent is only as powerful as the systems it can connect with via APIs and web hooks. — That is the crucial final piece. Let's go back to your software renewal example. You use the AI insight to negotiate the right size renewal down to 300 seats. The moment that new agreement is electronically signed via the DocYsine AM platform, the Docyign Iris engine extracts the new terms. The agent then automatically pushes that updated structured data directly into your outside spend management platform like Koopa for instance — without a human doing any manual data entry. — Exactly. It triggers Koopa to automatically generate a perfectly accurate purchase order for 300 seats updating your accounting systems and reflecting the new reality of the contract across your entire enterprise architecture. So synthesizing everything

we have discussed in this podcast, the AI is no longer just a fancy search engine or a grammar checker. It is an active participant in your business workflow. It is taking the tedious manual busy work entirely off the plate of your legal sales and procurement teams. It is allowing human beings to focus on strategy on highlevel negotiation and on building actual business relationships. — That is the ultimate realization of intelligent agreement management. The Docyign AM platform connects the entire journey. Data flows seamlessly from the moment of creation through the highly secure commitment phase into intelligent management and directly back into your core business systems via agents. — If you are listening to this, the way you work is fundamentally changing. Whether you are sitting in legal drowning in triage requests or in sales trying to get a perfectly compliant deal signed before the quarter ends or in procurement trying to track down millions in auto renewals, you are no longer buried in manual reviews or lost PDFs. By utilizing the DocYsine IM platform and Docuine Iris, you gain unprecedented speed, total operational control, and massive cost savings. — It really forces us to reconsider how we define what a contract actually is. For centuries, a contract was a static piece of physical paper. Over the last 20 years, it became a static piece of digital paper. But we are now entering an era where contracts are essentially active agentic software. Software that can monitor its own obligations, flag its own risks, and trigger its own renewals across external systems. And the question you really have to ask yourself is, when our agreements become self-managing software, how will that redefine the sheer speed at which global business can operate? — Thank you for joining us for this episode. We hope you enjoyed the discussion. Listen to another episode to learn more about the Docuign AM platform and how to unlock the value of your agreements.