AI in Legal
Introducing Review 2.0: Contract Review That Knows What Your Team Has Agreed To
Written by
Last updated
Every contract review tool on the market reviews your contract against a static playbook. To do so, every clause is evaluated individually to see whether it aligns with a corresponding playbook position. Playbooks are formulaic and unchanging, and lack nuance and discretion as a result.
Today we’re releasing the next generation of Ivo Review. It turns the unnatural style of playbook agreement review on its head; it no longer discretely decides whether a clause matches a particular standard in a vacuum, but instead, applies your playbook across clauses in a logical, coordinated way that mirrors how an experienced attorney actually negotiates. Now Ivo Review can understand not only what your positions are, but how they work together, and discern when a position versus a fallback position makes sense within the context of the agreement.
That context matters so much in a negotiation. A liability cap that's standard in a low-risk, short-term vendor agreement might be completely unacceptable in a strategic multi-year partnership. With Review 2.0, you give the context to Ivo, and it will use that to inform its suggestions. It even identifies and surfaces specific language you’ve previously approved or accepted in similar deals, and provides market-based suggestions.
We wanted to know whether all of these improvements to Ivo Review made a difference in performance, so we ran an independent benchmarking study to assess the performance. We gave 19 commercial agreements to an experienced attorney at an Am Law 25 firm, Ivo Review 2.0, and a leading AI model to review, and asked three independent senior lawyers to judge their performance.
Review 2.0 significantly outperformed off-the-shelf AI and performed on par with the human lawyer, We also found that Review 2.0 did better than the others in surgical redlining and judgment. And finally, Review 2.0 vastly exceeded Review 1.0's capabilities, performing 41% better in overall accuracy.
For the first time, an AI can exercise legal judgment. This judgment comes from contract benchmarks, your playbooks, and the context of the agreement. To do this, we built a direct connection between every clause you review and your actual negotiation history. Every time your team runs a review, Ivo allows users to select applicable agreements in your repository to benchmark against. And within that set of agreements, Ivo automatically narrows it further to similar agreements based on factors like contract type, governing law, your role as a party, and the counterparty's industry. This means that for every clause you review, you can see exactly how it compares to what your team has historically agreed to. You know whether a position is standard for your organization or if it deviates and, if so, by how much. This data is grounded in your own executed agreements and your negotiation outcomes.
We also changed the playbook itself. In Review 1.0, every position had to be broken into formulaic checklist items. The AI had no discretion on what to surface or when. If a position was in the playbook, it flagged it, whether or not the context made it relevant.
Review 2.0 reads your playbook the way an experienced attorney would. It understands not just what your positions are, but when to apply them, which fallbacks to recommend, and when the deal context means a position simply doesn't apply. Teams managing playbooks across multiple jurisdictions can now work from a single global playbook. Review 2.0 identifies the applicable jurisdiction and applies the right position automatically. You can also generate playbook positions directly from your existing templates and historical agreements in minutes.
Finally, Review 2.0 has a capability that no playbook can capture. It understands the context of the deal itself. For example, a liability cap that's standard in a low-risk short-term vendor agreement might be completely unacceptable in a strategic multi-year partnership. Your playbook can't know which situation you’re in, and, until now, neither could your AI.
With Review 2.0, you define what the deal is before review begins, and that context shapes everything. What to push on and what to accept. The AI isn't just checking your contract against a list of positions. It's applying judgment informed by what this deal actually means to your business.
Review 2.0 assigns a separate AI agent to evaluate each topic in your agreement. A superior agent sits above them, reconciling conflicts and consolidating their findings. This allows Ivo to reason across your entire agreement holistically, rather than processing it item by item.
What you get isn't a wall of overlapping checklist items that still require a lawyer to sort through. You get a clean set of consolidated recommendations, organized by topic, with the reasoning behind each one.
Every contract your team signs is a decision about risk. And the quality of those decisions depends on the quality of the context behind them. Until now, that context has been fragmented, either scattered across static playbooks, buried in old agreements living in someone's hard drive or even a filing cabinet, or living in people's heads.
Review 2.0 changes that. Every review now carries the wisdom of everything your team has ever negotiated, surfaced automatically and applied intelligently at the moment it matters most. It's your playbook, shaped by your precedents, sharpened by deal context, and backed by the kind of judgment your best lawyer brings to every contract.
Book a demo today and see Review 2.0 in action.
Schedule a demo today.
