AI adoption without strategy: Why 78% of professionals use it and only 15% see results
78% of legal professionals use AI, but only 15% see results. Discover why strategy-free adoption fails and how to fix it.
There's a pattern we see repeating across very different sectors. A company adopts artificial intelligence tools. Employees use them. Executives believe they're already "doing something with AI." But when someone asks what concrete results have been generated, no one has a clear answer.
The legal sector is a particularly interesting case for understanding why this happens and what can be done about it.
The Numbers That Don't Add Up
According to Litify's State of AI in Legal report (2025), 78% of legal professionals already use generative AI at a personal level. In 2023, the figure was 23%. That's a huge jump in just two years.
But the same report reveals that fewer than 15% of organizations report any real business impact from that adoption. The Legal Industry Report 2026 from 8am, based on more than 1,300 professionals, confirms the trend: although 42% already use AI tools specifically for legal work (double the previous year), firm-wide adoption remains at 34%.
How is it possible that almost 8 out of 10 professionals use AI and that almost no organization sees measurable benefits?
Chatbots Aren't the Same as an AI Strategy
Much of the explanation lies in what tools are being used and how. The three most popular among legal professionals are ChatGPT (66%), Microsoft Copilot (42%), and Google Gemini (24%), according to Litify. All three are general-purpose chatbots, accessible with a free account.
These chatbots are useful for specific tasks. Summarizing a text, looking up a concept, generating a draft. But there's a huge difference between a professional using a chatbot on their own and an organization having an AI strategy integrated into its processes.
Applied artificial intelligence in a business goes far beyond a chat window. It includes workflow automation, document analysis, alert systems, specialized agents that execute complex tasks autonomously. Confusing "my team uses ChatGPT" with "we have an AI strategy" is the first mistake many organizations make.
The Shadow AI Problem
When professionals adopt tools on their own, without policies or oversight, what's known as "shadow AI" appears. Each person chooses their own tool, inputs whatever data they consider appropriate, and works without coordination with the rest of the team.
In the legal sector, the numbers are revealing. According to 8am, 43% of firms have no formal AI policy and don't even plan to create one. Only 9% have an active policy that's actually applied. Less than half of professionals receive sufficient training on these tools (Litify, 2025).
This isn't exclusive to the legal sector. We see it in consulting, financial services, operations. The pattern is the same: rapid individual adoption, slow organizational adoption, and a governance vacuum in between.
Where It Does Work (and What They Have in Common)
The Future Ready Lawyer 2026 report from Wolters Kluwer shows the other side. 92% of legal professionals already use at least one AI tool, and 62% report weekly savings of between 6% and 20%. More than half say they've seen similar revenue increases.
But these results aren't distributed uniformly. They concentrate in organizations that share three characteristics:
- They have clear policies. They define what tools can be used, with what data, and for what purposes.
- They train their teams. They don't assume people already know how to use these tools. They invest in practical training.
- They integrate AI into their workflows. They don't depend on each person finding their own solution. The tools are connected to existing systems and processes.
What Your Company Can Do Now
If your organization is in the "everyone uses what they can" phase, you're not alone. Most are there. But the leap from individual adoption to strategic adoption doesn't require a massive transformation. It requires taking concrete steps:
- Audit what's already being used. Ask your team what AI tools they're using. You'll probably be surprised.
- Define a minimum policy. You don't need a 50-page document. Just establish what type of data shouldn't be entered into general-purpose chatbots and what tools are approved.
- Identify the processes that would benefit most. Not all processes need AI. Start with those that are repetitive, time-consuming, and prone to frequent errors.
- Measure something. If you're not measuring impact, you can't know if it's working. Time saved, errors reduced, capacity freed up.
The Difference Is in Management, Not Technology
The gap between the 78% who use AI and the 15% who see results isn't a technical problem. It's a management problem. The tools exist and are increasingly accessible. What's missing in most organizations is the structure to use them with criteria.
The good news is that closing that gap doesn't require large investments. It requires clarity about what you want to achieve, willingness to measure results, and often a partner who understands both the technology and your operations.
Sources:
- Litify, State of AI in Legal Report (2025). litify.com
- 8am, Legal Industry Report 2026. 8am.com
- Wolters Kluwer, Future Ready Lawyer Survey Report 2026. wolterskluwer.com