Daniel Koren is CEO and founder at StartWise.
The tech industry used to depend on niche technical knowledge for exclusivity. But in the last year, there’s been a shift toward “vibe coding,” as OpenAI cofounder Andrej Karpathy calls it. While I think this is a fun phrase, I look at vibe coding as a new way to build things that feels less like you’re translating specifications to syntax, and more like actually sculpting an idea in real time.
Instead of weeks of writing boilerplate, my teams are now “vibing out” early concepts, and AI is doing the heavy lifting. We’ve gone from an idea spark to a working prototype in hours. While this sped-up process is exhilarating, it also has real consequences when not consciously managed.
What Is Vibe Coding, And Why Is It Trending Now?
The traditional software development process has always been a grind: You write specs, translate them into code, debug and then ship. Vibe coding flips this process on its head. You simply describe what you want in plain language, and then AI drafts the scaffolding for you.
The magic isn’t in the automation; it’s in the flow. Instead of agonizing over every edge case up front, you can focus on speed, alignment and momentum. For startups, this means you’re not leaving ideas to die in planning decks. For teams, this means prototypes get on the table in days. For investors, this means improving capital efficiency as timelines collapse.
Why is vibe coding trending now? Because three forces have converged:
1. AI models have matured enough to generate usable, production-grade code.
2. Market competition punishes slow movers.
3. We’re seeing a new wave of founders who don’t come from engineering backgrounds, but they can now build.
How Vibe Coding Delivers In The Real World
Vibe coding is making huge changes in different industries. I’ve seen it firsthand. During my time with a gaming startup, we were able to develop a playable prototype of a new concept in two days that previously would have taken us weeks to develop.
Marketing teams that once relied on engineering teams are developing their own tools. I’d say that thousands of minimum viable products are being created in weeks now instead of quarters.
I’ve seen this across many startups: rapid prototypes, greater velocity in decision making and an overall change in momentum that is contagious across teams. The common denominator is clear: faster cycles, tighter alignment and a different energy that creates future alignment and momentum.
A simple example: landing pages for marketing. It used to take our team one to two weeks to create a high-quality page. But after building a knowledge base that captures everything we care about—like user interface design, key data points, lead capture mechanics, and even chat and voice agents for lead capture and closing the deal—we’ve cut that time down to just one to two days, and with better quality and results.
Behind The Movement
The Model Evaluation & Threat Research (METR) organization recently studied the impact of AI tools on software development and found that using them slowed down experienced developers. I’ve seen this myself: People feel like they are moving faster, but feelings are not reality.
On the other hand, IBM has highlighted how teams can use vibe coding to streamline processes and accelerate innovation. That aligns with my experiences: When teams develop shared practices, adoption goes fast. That’s why processes and review are important; there is no negotiating.
I’ve watched leading enterprises doing experiments with vibe coding, but the real breakthroughs don’t come from which platform you choose; they come from how you view and use the technology. The companies winning in this particular gap are the ones that are collapsing developmental cycles, aligning teams and treating AI as a collaborator rather than a black box.
Addressing The Risks And Challenges
Just like any other major shift, vibe coding carries a major set of vulnerabilities that must be assessed right away.
Security Flaws
Security flaws remain one of the most pressing concerns. AI-generated code can easily pass the functional tests while quietly neglecting and introducing weaknesses.
Maintenance Needs
Without a careful human review, AI outputs can create a technical debt. Code that works now could become obsolete in future updates.
Skill Atrophy
Skill atrophy could also be a problem. Overreliance on AI could potentially diminish developers’ ability to debug or architect systems independently, particularly among juniors.
Ethical Risks
AI models can unintentionally replicate the biases embedded in their training data, which can lead to discriminatory behavior in applications.
The consensus among experts is not to dismiss the potential of vibe coding, but to integrate it into a hybrid approach with human-driven architecture, AI-facilitated implementation and strict code reviews.
Adopting Vibe Coding
If you’re thinking about experimenting with vibe coding, my advice is to start with a small project to test the workflow and get a feel for how everything fits together. Prioritize iteration speed over perfection; moving fast and learning as you go will get you further than trying to get everything right up front.
Keep humans in the loop to review and secure AI-generated code, ensuring quality and safety. And make sure your whole team is working in shared environments so alignment happens in real time.
Investors interested in vibe coding should bet on teams that rely on velocity and governance. CEOs should rethink talent plans not to replace engineers, but to elevate them into system architects who can guide AI-driven development.
From my experience in working with emerging AI development teams, I’m convinced that vibe coding is not just a trend, but rather a structural change. The organizations that will create the most meaningful value are not those simply chasing speed, but those that know how to marry the creative speed of AI with best practices, using it to deliver both faster and with higher quality.
Forbes Agency Council is an invitation-only community for executives in successful public relations, media strategy, creative and advertising agencies. Do I qualify?
link