BlackBoiler has unveiled Veris, a new contract-review platform that combines its established deterministic editing engine with generative AI and an agentic chat interface.
The company has spent over a decade developing automated redlining technology, and Veris represents its most significant product evolution to date.
Veris runs directly inside a Microsoft Word add-in, allowing contract-review teams to negotiate and mark up agreements without ever leaving the document.
Alongside the launch, BlackBoiler is introducing two new subscription tiers designed to broaden its reach beyond its traditional enterprise customer base.
The Starter tier, aimed at solo reviewers, is priced at $1,250 per year, while the Pro tier, built for recurring team reviews, costs $3,000 per user per year.
Daniel Broderick, co-founder and CEO of BlackBoiler, said Veris was developed in direct response to consistent customer feedback about wanting a more interactive experience.
“They wanted that in a more agentic experience with faster onboarding,” he said.
BlackBoiler’s original system predates the current generation of large language models, using an organisation’s historical contract markup data to drive its edits rather than predictive text alone.
Veris retains that deterministic foundation while incorporating large language models “where necessary,” with strict controls over what data is shared with those models.
The platform includes a robust validation layer that statistically analyses every suggested edit, measuring how dramatically a sentence changes and tracking specific word additions or deletions.
That validation process cross-references suggested changes against similar edits BlackBoiler has processed previously, with the explicit goal of limiting hallucinations across all outputs.
Broderick said the system is designed to “only edit what needs to be edited and not more than what needs to be edited,” reflecting the platform’s core philosophy around precision.
Robert Moore, BlackBoiler’s director of sales, said the deterministic approach is what separates Veris from simply using a large language model on its own.
A “judge” component validates each suggested edit by tracing it back to the specific examples a customer originally provided, ensuring consistency with a company’s own negotiating standards.
One of Veris’s headline features is its ability to automate the building of a playbook, the master ruleset that governs how an organisation wants its contracts edited.
Broderick demonstrated building a playbook by uploading a single marked-up contract, with the system extracting roughly 20 rules from that document alone.
Users can also create playbooks by uploading policy documents, written risk descriptions, or simply by describing rules directly through the chat interface.
Veris then runs an “enhancement loop” in the background, generating prompts, searching for similar clauses, applying edits, evaluating results, and refining both prompts and judges automatically.
Broderick said this removes the human variable from prompt engineering entirely, because “the data should build the prompts” rather than individual lawyers writing their own versions.
Veris offers two distinct review modes, with a “full review” inserting all suggested edits as tracked changes and a “quick review” placing revisions in the margin for users to accept individually.
“Instead of relying on each user’s prompting skill, Veris derives prompting standards from the edits and review behaviors that define how an organization negotiates,” Broderick said.
Broderick believes the platform signals where the broader contract AI industry is heading, arguing that governance and consistency will define the next phase of adoption.
“The next phase of contract AI will be shaped by consistency, governance, and cost-efficient execution,” he said, “not just language generation.”