Control is All You Need: Why Most AI Systems & Agents Fail in the Real World, and How to Fix It
Last Updated on September 9, 2025 by Editorial Team
Author(s): Kenny Vaneetvelde
Originally published on Towards AI.
Control is All You Need: Why Most AI Systems & Agents Fail in the Real World, and How to Fix It
If you’ve spent any time on social media the past few months, you’ve been sold a very specific dream about AI agents. You’ve seen the demos: an “autonomous” agent that builds a complete Snake game from a single prompt (Wowzers!), or an entire “research crew” that generates an amazing comprehensive report on a complex topic that in reality some times performs wonderfully, some times outputs trash, and more often than not doesn’t give you 100% exactly what you were expecting. There is this compelling vision of digital employees handling complex tasks with little to no human intervention.
The article critiques the current excitement surrounding AI systems and agents, arguing that they often fail to deliver reliable results in real-world applications due to their reliance on a ‘black box’ approach, leading to unpredictable performance and costs. It emphasizes the importance of returning to established software engineering principles, promoting modularity, clear interfaces, and well-defined schemas to manage AI capabilities effectively. The author suggests a more controlled integration of AI into existing workflows to ensure reliability and accountability, ultimately advocating for building practical systems rather than merely chasing flashy demos.
Read the full blog for free on Medium.
Join thousands of data leaders on the AI newsletter. Join over 80,000 subscribers and keep up to date with the latest developments in AI. From research to projects and ideas. If you are building an AI startup, an AI-related product, or a service, we invite you to consider becoming a sponsor.
Published via Towards AI
Towards AI Academy
We Build Enterprise-Grade AI. We'll Teach You to Master It Too.
15 engineers. 100,000+ students. Towards AI Academy teaches what actually survives production.
Start free — no commitment:
→ 6-Day Agentic AI Engineering Email Guide — one practical lesson per day
→ Agents Architecture Cheatsheet — 3 years of architecture decisions in 6 pages
Our courses:
→ AI Engineering Certification — 90+ lessons from project selection to deployed product. The most comprehensive practical LLM course out there.
→ Agent Engineering Course — Hands on with production agent architectures, memory, routing, and eval frameworks — built from real enterprise engagements.
→ AI for Work — Understand, evaluate, and apply AI for complex work tasks.
Note: Article content contains the views of the contributing authors and not Towards AI.
Related posts
Recent Posts
