Legal questions are among the first things people try on an AI chatbot, and among the last things they should trust it with. The answer comes back fast, fluent, and with nothing behind it, which is fine until the question is whether your contract covers a client's breach. AI can do real work in legal operations, but only when the answer is grounded in your own agreements, traceable to a source, and handed to a lawyer when it stops being clear.
Rob Scott, an attorney with more than 25 years in the managed IT services channel, has built his practice around that standard. Scott is also the co-founder of Monjur, which delivers attorney-supervised contract intelligence exclusively for managed service providers (MSPs). Built on the law firm Scott & Scott, LLP, Monjur writes and maintains the contracts that protect more than 1,000 MSPs.
How AI is improving speed and accuracy in contract work
One of the capabilities of AI that many lawyers have leveraged is its capacity for advanced data analytics. In the legal profession, data analytics are key to allowing lawyers to draft and review contracts. An AI model can be set up to identify and recommend attorney-approved clauses for a contract draft or identify any abnormal language that needs a lawyer’s particular attention. Because of this, AI is an essential step towards greater efficiency and accuracy in contract drafting and review.
“At Monjur, our attorneys set the guardrails and Pilot, our legal AI assistant, works inside them, which frees the team for the higher-value work of strategic negotiation and risk assessment,” explains Scott. “The system flags non-standard language against clauses our attorneys have already approved, so issues surface early and nothing gets invented on the fly. An MSP owner gets an answer in seconds instead of waiting a week for a callback.”
Furthermore, considering the repetitive nature of document review and contract drafting, it is an error-prone process, even for the most experienced legal professionals. AI tools do not deal with issues like fatigue or distraction, allowing them to remove some of the most common sources of human error in the document review process. As a result, lawyers can improve not only their speed but also their consistency with this monotonous process.
“Pilot answers contract questions grounded only in that MSP’s own agreements and our attorneys' prior decisions, and it cites the exact source for every answer,” says Scott. “When its confidence drops below our threshold, it does not guess. It stops and routes the question to an attorney. That is the line between a legal tool and a chatbot.”
Indeed, consistency is key in regulatory and internal compliance, where missing even the smallest detail could become a costly mistake. Thankfully, automation and natural language processing (NLP) technologies will consistently identify areas of concern based on their programming and training, making AI a powerful resource for lawyers looking to streamline compliance reviews.
“Our attorneys track the regulations our clients operate under, from state privacy laws to CMMC. The terms live at a link the agreement points to, so when something changes, we update that language once and it reaches every agreement pointing to it, with no need for anyone to re-sign. Each client chooses whether an update goes live automatically or only after they review it,” Scott explains. “The technology is what makes that maintenance affordable. The judgment about what needs to change is still an attorney's.”
How AI helps lawyers assess and reduce risk
Some lawyers have even taken their integration of AI technology a step further by using artificial intelligence’s data-driven insights to inform decision-making. “Rather than training a model on open data, we ground Pilot in a closed knowledge base built for a single MSP: their own agreements, the rules that apply to them, and the record of decisions our attorneys have already approved for them,” says Scott.
Similarly, this technology can be used to conduct risk assessments, identifying and analyzing high-risk clauses for lawyers. This allows them to develop a solution to protect themselves and their clients, which could prove instrumental in resolving disputes before they escalate.
Scott recalls an instance in which his team reviewed an MSP’s vendor agreements with AI support. It surfaced inconsistencies in key indemnification clauses, prompting them to rewrite how that risk passed through to the client's own customers. Because an MSP assembles its service from tools it does not own, a vendor outage or breach lands on the MSP unless the paper passes that risk through. The concern would have surfaced in a lawyer’s review either way. The AI put it on the desk immediately.
“The risk I see is rarely a badly drafted agreement. It is an agreement that was accurate the day it was signed and has not kept up,” Scott adds. “A service line gets added, an engineer swaps a backup product, a client starts using AI tools. Nothing in the business flags any of that, and the paper sits in a folder looking as valid as it ever did. What the technology changes is how often somebody looks.”
For a business weighing whether to let AI near its legal work, the question worth asking is what happens when it gets answers wrong. “Clients notice the speed,” Scott says. “The reason they trust the answer is that a licensed attorney stands behind it, and the system knows when to go get one.”
