The era of passive AI chatbots is rapidly shifting toward active digital workers. Stockholm startup Kovant has successfully raised €1.5 million in pre-seed funding to build fully autonomous AI agent workforces for enterprises. This investment marks a significant step away from simple co-pilots and moves toward systems that can manage complex business operations with minimal human interference.
The Rise of Agentic Workforces
Businesses today are overwhelmed with “AI co-pilots” that still require constant human prompting. Kovant aims to solve this fatigue by introducing agentic AI. These are not just chat bots that answer questions. They are digital workers capable of executing tasks.
Most companies remain stuck in endless proof-of-concept loops with their AI tools. They struggle to see real value because the software requires too much hand holding. Kovant changes this dynamic completely.
The platform allows businesses to deploy entire departments of AI agents. These digital teams handle specific functions like supply chain management, procurement and customer success. They do not wait for a prompt. They act based on set goals and only ask for help when necessary.
This approach mirrors a human workforce structure. You have specialized workers handling execution while managers provide oversight. Ali Sarrafi, the CEO and co-founder of Kovant, emphasized a crucial shift in their strategy.
He believes that AI has recently stripped the human element out of work by forcing people to manage bots. Kovant flips this script by letting AI handle the execution so humans can focus purely on judgment and strategy.
autonomous enterprise ai agent swarm network visualization
Smart Tech Reduces AI Hallucinations
A major fear for any enterprise adopting AI is accuracy. Large Language Models (LLMs) are powerful but often prone to “hallucinations” or making up facts. This risk makes big companies hesitant to trust AI with critical financial or operational data.
Kovant tackles this problem with a unique technical architecture. They do not rely solely on massive, general purpose models. Instead, they utilize swarms of agents based on Small Language Models (SLMs).
Why SLMs are safer for business:
- Focused Knowledge: They are trained on specific tasks rather than the entire internet.
- Lower Error Rate: Reduced complexity leads to fewer hallucinations.
- Better Compliance: It is easier to set strict guardrails for auditing.
- Speed: Smaller models process data faster and cost less to run.
This “swarm” intelligence allows different agents to check each other’s work. One agent might draft a procurement order while another verifies it against compliance rules. This internal check ensures high accuracy before a human ever sees the result.
Turning Month Long Tasks into Days
The true measure of any business tool is the impact it has on the Profit and Loss (P&L) statement. Kovant is already showing tangible results with its early clients. The company focuses on heavy operational tasks that usually drain human resources.
Consider the procurement process.
In a typical enterprise, managing bids and procurement can take up to a month. It involves endless emails, compliance checks and data entry. Kovant’s portfolio shows these processes dropping drastically in duration.
“We are seeing autonomous executions completed within one to two days while retaining human oversight.”
This efficiency allows companies to scale without hiring more administrative staff. The AI works continuously in the background. It learns over time and gets faster.
Below is a comparison of the traditional workflow versus the Kovant approach:
| Feature | Traditional Co-Pilot | Kovant Agentic Workforce |
|---|---|---|
| Trigger | Human must prompt every step | Autonomous action based on goals |
| Scope | Single task assistance | End-to-end process management |
| Oversight | Constant supervision needed | Review only on escalation |
| Speed | Limited by human speed | 24/7 continuous operation |
The system is designed for rapid deployment. Enterprises can get these agent teams running in weeks rather than months. This speed to value is critical in the current economic climate.
Strategic Investors Fuel Expansion Plans
The €1.5 million funding round attracted high profile backers who understand the enterprise landscape. Stockholm based VC firm J12 Ventures led the round. They were joined by Ampli, Green Ventures and several key angel investors.
Notable among the investors is Sara Öhrvall. She serves as a board member at major companies like Axfood and Verisure. Her decision to join the Kovant board signals strong industry confidence in the startup’s vision. Emmanuel Martin-Chave, the VP of AI at Data Guard, also participated in the round.
This capital will fuel specific growth targets:
- Geographic Expansion: Kovant plans to enter markets in the Nordics, Switzerland and the Benelux region.
- Product Development: Further refinement of the SLM agent swarms.
- Global Reach: Making the platform available to enterprises of all sizes worldwide.
The backing by industry veterans suggests that the market is ready for “Agentic AI.” Companies are looking for tools that do the work rather than just talking about it.
Kovant represents a maturing of the artificial intelligence sector. We are moving from the novelty phase of chatting with computers to the utility phase of computers running operations. With solid financial backing and a clear focus on solving real business pain points, Kovant is well positioned to lead this shift. The ability to reduce month long tasks to mere days provides an undeniable competitive advantage for their clients.
What are your thoughts on trusting AI agents to run business operations autonomously? Share your opinion in the comments below. If you are excited about this shift in technology, please share this article on social media using #AgenticAI and #Kovant.
