November 2025

AI: experimenting with confidence.

AI has everyone talking, but for most legal teams, it’s still more curiosity than capability.

In a recent Front Foot Forward webinar, Sarah Gaidzkar (GC, PageUp) and Horace Wu (Founder & CEO, Syntheia) shared what it really takes to experiment with AI confidently, and what separates progress from noise.

Here are the six key takeaways.

1. Be clear on your goals

Start with a problem, not a tool. Define what success looks like before you start testing. FOMO isn’t a strategy; clarity is. If the problem or the goals aren’t clear, the outcomes won’t be either.

2. Keep humans in the loop

AI should support, not replace, legal reasoning. Lawyers still make the calls, deciding what to use, what to question, and what to discard. AI can assist, but accountability stays with people.

3. Measure performance and ROI
Treat every AI trial like an experiment. Set simple success criteria: speed, accuracy, usability, and cost. Compare results against your current baseline. If it adds measurable value, keep going. If not, move on and adjust.

4. Strengthen the foundations
AI can’t fix weak systems. Solid document management, good knowledge sharing, contract processes, and data governance matter more than ever. Without them, even the best tools will struggle to deliver results.

5. Address the team first
Many challenges aren’t technical; they’re human. Role confusion, unclear priorities, and misalignment hold teams back far more than a lack of technology ever will.

6. Know when to push and when to pause
If AI is showing strong ROI, build on it. If not, get back to fundamentals. Experimentation should move the needle, not distract from what already works.

Confident innovation doesn’t come from chasing the latest tools; it comes from purpose, structure, and informed human judgment.

Because the smartest teams aren’t just doing more with AI, they’re doing the right things better.

You can watch the full discussion below.