The most important AI capability going mainstream in the next 12 months
The biggest shift will be AI moving from “help me do this task faster” to “help me accomplish my intent”.
A lot of the AI adoption so far has been useful, but fairly narrow. Summarize this. Draft that. Automate this workflow. Those things matter, but they are mostly scaling human effort. The more interesting capability is when AI starts helping companies see patterns they would not have seen on their own and make better choices because of it.
You can see this clearly in audience development. Historically, a business had to decide who it thought the right audience was to market to. With AI, you can start to work backward from the outcome you want and let the system help identify who is actually most likely to respond. That is a very different way to think about using AI. It moves companies from assumption-based inputs to outcome-based recommendations.
How AI changes organizational decision-making by 2027
By 2027, I think the best organizations will make decisions with better inputs and a lot less guesswork.
That doesn’t mean AI replaces leadership or judgment. I do not believe that. But most companies still make too many decisions based on habit, instinct, incomplete information, or whoever made the most convincing argument in the room. AI gives teams a way to process more signals, challenge assumptions, and understand what is actually happening faster.
In marketing, that shows up in the gap between who a business thinks it should be targeting and who is actually likely to take action. That same pattern exists in a lot of decision-making. You have an assumption, you have the available data, and then you have what the data actually proves. AI should help close that gap faster. The companies that use it well will not just automate decisions. They will make better ones.
The most underestimated AI risk / bet leaders will get wrong
Bad data.
There is a lot of attention on the models, the tools, and the use cases, but not enough attention on the quality of the data underneath them. If the data is stale, incomplete, or wrong, AI can give you a very confident answer that is still completely wrong.
A false positive is one of the most damaging things AI can produce because it gives the illusion of certainty. Companies need to be much more serious about data accuracy, freshness, depth, and breadth. Otherwise, AI just becomes a more efficient way to be wrong.
Advice for a company starting its AI journey today
Start with the business problem. Do not start with “we need to use AI.”
That sounds obvious, but a lot of companies are doing the opposite. They feel pressure to show they are doing something with AI, so they start experimenting without a clear idea of what they are trying to improve. Are you trying to move faster? Reduce waste? Improve customer experience? Make better decisions? Build something competitors can’t easily copy? Those are different goals, and they require different approaches.
I would also avoid getting stuck in endless exploration. Give yourself a defined window to learn, test, and benchmark, then start making decisions. You can hire and partner at the same time. You do not need to wait until the perfect team or perfect governance model exists before you start.
The AI trend you are most excited about
I’m most excited about AI moving beyond productivity and into decision intelligence.
The first wave was mostly about speed. Write faster. Summarize faster. Analyze faster. That’s helpful, but the bigger opportunity is using AI to find patterns, recommend actions, and help businesses make better decisions than they could with human analysis alone.
For us, that shows up in AI audiences and predictive targeting. Instead of asking, “Who do we think we should target?” AI can help answer, “Who is actually most likely to take action?” That’s a much better question.
More broadly, I am excited about AI that helps companies see around corners. Not in a magic way, but in a practical way: better signals, better timing, better decisions, and systems that continue to learn. That’s when AI stops being just another productivity tool and starts becoming part of the company’s competitive advantage.
