The AI market right now feels overwhelming, and that is by design.
There is a new tool being pitched to law firms every week that’s promising to change everything.
Most firm owners have tuned it out entirely, which was the rational response.
The problem is that tuning out the noise also means missing the actual decision that matters. And that’s which platform your firm is going to build on.
We spent real time working through this at Sterling Lawyers.
Our team landed on Anthropic’s Claude Enterprise as our platform of choice.
The reasoning behind that decision is worth sharing, less because I think every firm should make the same call and more because the framework we used to get there applies regardless of which provider you end up with.
The Real Field of Options
There are four platforms worth considering at this point — Anthropic’s Claude, OpenAI’s ChatGPT, Google’s Gemini and xAI’s Grok.
When our team went through this exercise, Grok was not yet built out enough at the platform layer to be a serious contender for enterprise use. That left three real options for evaluation.
The three criteria our team weighed were terms of service, quality of the user interface for non-technical professionals, and depth of accessible capabilities for those same users.
Terms of service came first because everything else is irrelevant if the contract itself does not protect client data.
The interface question mattered because the people using this every day inside a firm are attorneys, paralegals, intake staff, and administrators, not developers.
If the tool is too complex for the average professional to use comfortably, adoption dies.
Claude came out ahead on all three when our team ran the evaluation.
The user interface is genuinely friendlier for non-technical professionals.
The agent capabilities were further along at the time we chose.
The terms of service structure worked cleanly for how we needed to operate.
That gap has narrowed on some fronts since. OpenAI has closed the distance on the agent side, but the overall fit still holds for us.
The Long-Term Wildcards
One thing worth understanding is that the current leader may not be the eventual winner.
Our team’s read is that Google and xAI both have significant long-term advantages tied to infrastructure that could reshape the field in a few years.
Google is deeply vertically integrated with its own chips and effectively unlimited cash flow to invest.
xAI has a compute story that no one else can match if it plays out the way it is being built.
For firm owners, the practical takeaway is that whichever platform you build on today is a decision that may need to be revisited. That’s fine.
The learnings transfer. The skills and workflows you build on one platform migrate reasonably well to another.
What you cannot afford is to keep waiting for a perfect platform to reveal itself. There is no perfect platform.
There is one you commit to, train your team on, and integrate into how the firm operates.
Why One Platform Instead of Nine
The other decision worth flagging is going deep on one platform rather than spreading across many.
A common pattern I see is firms with nine different AI subscriptions. One tool for document drafting, one for intake, one for calendar summaries, one for research, and so on.
Each one is a separate contract, a separate login, a separate learning curve for the team, and a separate governance headache.
Consolidating around a single platform simplifies almost every part of this. Governance becomes manageable. Training becomes coherent.
The productivity gains compound because the skills your team builds on one workflow transfer to the next.
And the cost picture is usually significantly better than paying for multiple overlapping subscriptions.
Adoption Is the Real Battle
At Sterling, weekly live building sessions, team breakout workshops, and one-on-ones have proved far more effective than policy documents or training decks. Leading and enabling staff to see real-time results motivates adoption.
Consequently, internal AI applications have automated heavy administrative tasks, allowing us to restructure or remove roles, shift our hiring pipeline, and improve firm margins.
The Point of All of This
If you are a firm owner sitting on this decision, the framework is straightforward. Evaluate the terms of service first. Weigh the interface heavily because your team has to actually use it. Look at the depth of capabilities for your specific use cases.
Then commit and start building.
The gap between firms that have made this call and started building versus firms still evaluating is going to keep widening. Waiting for perfect information is a slower path than picking a strong option and going.
