The world of artificial intelligence is currently at a critical transition point. While 2023 and 2024 were all about "chatting" with models like ChatGPT or Claude, 2025 and 2026 belong to autonomous agents. Capita, which manages processes for massive corporations and public administration, is not waiting for this trend — it is actively shaping it.
What exactly is "Agentic AI"?
To understand the context, it's necessary to distinguish between standard generative AI (GenAI) and agentic AI. While a standard model (e.g., GPT-4 or Gemini) works as a very smart text generator that answers your query, agentic AI has the ability to plan, use tools, and execute steps to achieve a goal.
Imagine the difference:
- GenAI: You type: "Book me a flight to London." The model writes you an itinerary.
- Agentic AI: You say: "Book me a flight to London." The agent logs into your account, compares prices, selects the best connection based on your preferences, makes the payment, and confirms the reservation in your calendar.
Strategic pivot: From contact centres to intelligent infrastructure
Under the leadership of new CEO Adolf Hernanz and Chief AI & Product Officer Sameer Vuyyuru, both of whom come from Amazon Web Services (AWS), Capita is radically changing its business model. The company has decided to sell part of its private sector operations (contact centres) in order to fully focus on segments with high levels of regulation and complexity — namely public administration and critical infrastructure.
This is a smart move. In these sectors, the priority is not a "quick response" but rather accuracy, security, and regulatory compliance. This is exactly where the strength of the partnership with Snowflake comes into play. The implementation of the Snowflake Intelligence system allows agents to work directly on data under strict control, which is key to maintaining data sovereignty.
Comparison: Who is leading the battle for agentic systems?
With this move, Capita positions itself in direct competition with other tech giants offering similar "agentic" capabilities:
| Platform | Primary focus | Strengths |
|---|---|---|
| Snowflake Intelligence | Data-centric AI | Direct integration within the data warehouse, high security. |
| Salesforce Agentforce | CRM & Customer Service | Extreme focus on sales and customer care. |
| Microsoft Copilot Studio | Enterprise Workflow | Deep integration into the Office 365 and Azure ecosystem. |
Why is this important for the telecom sector?
Capita is trying to attract large telecom operators (telco) such as Tesco Mobile or Virgin Media O2. For these companies, agentic AI represents a huge opportunity in the area of customer support and network management. An agent can autonomously resolve technical faults, optimize network capacities, or carry out complex billing corrections without the need for immediate human operator intervention.
At the same time, the question of data sovereignty arises. In an era where data is the new oil, telco companies must know where their data is located and who has access to it. Capita's "intelligent infrastructure" strategy targets precisely this solution — AI that runs in a controlled environment.
Impact on the Czech market and the European scene
Although Capita is a British company, its direction has direct implications for Czech companies and the technology sector:
- EU AI Act: The European regulation on artificial intelligence places great emphasis on transparency and security for "high-risk" systems. Autonomous agents that can make decisions about services for citizens will be under strict supervision. Companies in the Czech Republic that use outsourcing must require proof of compliance with these rules from their partners.
- Technology availability: Tools like Snowflake Intelligence are available to global enterprise clients, which includes large Czech corporations and state institutions.
- Pricing policy: Unlike standard B2C tools (e.g., ChatGPT Plus for $20/month), systems like Snowflake Intelligence are sold on an enterprise contract basis. The price is not public but is derived from the volume of processed data and the number of agentic actions. For the Czech market, this means that implementation will be costly, but with a high return on investment (ROI) thanks to process automation.
For a Czech user or small business, this does not yet mean a direct change in everyday life, but for medium and large enterprises in the Czech Republic, it is a warning: The future of outsourcing lies not in human hands-off processes, but in managing autonomous agents.
How is a chatbot different from an AI agent?
A chatbot primarily responds to questions based on provided information (conversational). An AI agent can use tools, access databases, and execute multi-step tasks autonomously to achieve a specific goal (actionable).
Is agentic AI safe for sensitive data under GDPR?
Security depends on the infrastructure. Platforms like Snowflake Intelligence are designed to run agents within a company's existing security protocols, thereby minimizing the risk of data leakage outside the regulated environment.
Can agentic AI replace people at work?
The goal is not full replacement, but the automation of repetitive and complex administrative tasks. People will thus need to shift into the role of "supervisors" or "orchestrators" of AI agents, overseeing their accuracy.
