Adi Prakash is Founder and CEO of Sentient Ventures, an AI-native firm helping companies scale smarter with AI as the multiplier.
We’re only a few months into 2025, but it already feels like the ground has shifted beneath us. For business leaders, the pace of change in artificial intelligence isn’t just fast—it’s exponential. And if it feels like the rules of the game are being rewritten in real time, that’s because they are.
What we’ve seen so far this year is a decisive pivot from experimentation to execution. AI is no longer about what’s possible. It’s about what’s necessary—and how fast you can deploy it.
Tectonic Shifts In AI Adoption
Over the past five years, we’ve experienced two separate “order-of-magnitude” leaps in AI capability. That means a 100x improvement—not once, but twice—in how well machines can perform tasks we once thought were uniquely human.
The first leap came with the rise of generative models. Suddenly, AI could write, summarize, reason, and synthesize language in ways that felt uncannily human. The second came with the collapse in cost and complexity. What used to require elite research teams and billion-dollar infrastructure is now available through open-source models and easy-to-integrate platforms.
Now, we’re seeing a third wave emerge: agentic AI. These are systems that don’t just assist, but act. They are goal-oriented, autonomous agents capable of executing tasks end-to-end across workflows. In Q1 2025 alone, mentions of “agentic AI” by CEOs spiked by 275%, a sign that business leaders are beginning to grasp the magnitude of this shift. When software can complete processes—book travel, schedule logistics, reconcile invoices—without waiting for human input, the entire nature of operational scale and decision velocity changes.
Under Pressure
Economic and geopolitical disruption has accelerated the AI evolution. In Q1 alone, U.S. tariffs on imports from China, Canada, Mexico and the EU reshaped cost structures and supply chains across multiple industries. Add rising energy prices, constrained labor markets and ongoing global uncertainty, and the picture is clear: the old playbook doesn’t work anymore.
These pressures are forcing companies to move faster on AI adoption. But more importantly, they’re changing how businesses use AI. It’s no longer about automating tasks or optimizing operations at the margins. It’s about redesigning your business for resilience.
Manufacturers, for example, are using AI to localize production, simulate tariff-induced shifts and make just-in-time decisions based on live customer demand. Retailers are automating supply chain routing and rethinking pricing models. Even the most risk-averse sectors—like finance and healthcare—are moving toward AI-native workflows.
Rewriting The Operating Model
The companies pulling ahead aren’t bolting AI onto legacy systems. They’re starting with a blank sheet of paper and instead of asking how to engineer AI into their current infrastructure, they’re asking a better question: What would our business look like if it had been built with AI from day one?
For example, let’s take a visit to the doctor. You check in with a receptionist or kiosk, endure a long wait for what amounts to a brief visit, then your prescription or treatment orders get faxed or emailed out.
Now, let’s take that journey reimagined with AI. Your phone app schedules and updates appointments in real time. A smart vestibule checks your vitals automatically. Your provider enters with your data already synthesized. The visit is focused, fast and integrated. Prescriptions are sent and filled automatically—maybe even delivered by drone.
That’s not a tweak to the current system. That’s a reinvention. And if it works for one clinic, one doctor, one patient—it scales.
This kind of thinking is showing up everywhere. In finance, copilots are giving way to autonomous financial agents. In logistics, companies are ditching static forecasts for dynamic, AI-led planning. In operations-heavy environments, agentic AI is managing multistep processes—everything from procurement to onboarding—with minimal human intervention. These aren’t efficiency gains. They’re strategic redesigns.
The Future Of Human Expertise
I get asked a lot: if AI can do so much, what’s left for people?
It’s a fair question. We’re clearly heading into a world where many forms of expertise—especially those based on repeatable knowledge—will be replicated by machines. But that doesn’t mean human value disappears. It shifts.
The real differentiators will be emotional intelligence, adaptability and conceptual thinking. Not “can I code?” but “which tool do I use, when, and why?” In other words, we’ll need fewer people who know everything, and more people who can orchestrate AI to solve the right problems.
We’ll also see the rise of small language models (SLMs) trained on highly specific company or industry data. That’s where a lot of the next wave of value will come from. And that means leaders will need to build internal AI fluency, not just outsource it.
The Middle Market’s Moment
Here’s what often gets overlooked: most of the action in AI has been concentrated among Fortune 500 companies with deep pockets and deep benches. I know—I worked with many of them at AWS.
But the real transformation needs to happen in the middle market. These companies face the same pressures but often lack the capital, technical muscle, or access to elite consultants. The good news is: that’s changing. With tools like Cohere and other business-tuned models, the barriers to entry are falling.
For the first time, small and mid-sized firms have a chance to leapfrog their legacy constraints. The playing field isn’t just leveling—it’s tilting in favor of those nimble enough to rethink their models and move fast.
The Real ROI Isn’t Flashy
The best returns we’re seeing in 2025 aren’t coming from flashy GenAI pilots or viral demos. They’re emerging in the backend—in the unsexy, operational guts of the business. Predictive maintenance, customer support automation, and white-collar task acceleration are consistently delivering 10x ROI.
But the real edge is in AI-led decisioning. Systems that don’t just recommend actions, but take them. These systems are enabling companies to respond faster, iterate smarter, and adapt in real time. And in a world where volatility is the norm, speed of decision becomes your biggest competitive advantage. If you’re still waiting for things to stabilize so you can figure out your AI strategy, you’re already behind.
Forbes Business Council is the foremost growth and networking organization for business owners and leaders. Do I qualify?
Read the full article here
