Companies Must Prepare for the Arrival of AI Agents
brussel, zaterdag, 19 juli 2025.
AI agents are advancing in the business world, but companies need to prepare well for the challenges that this technology brings. According to recent research, 70% of companies expect AI agents to play a significant role in their business processes within three years. Training and new procedures are crucial to ensure the smooth integration of AI agents and to fully utilise the potential benefits.
Training and New Procedures Crucial for Integration
To effectively handle the arrival of AI agents, companies need thorough preparation. Training employees and developing new procedures are critical steps. TechInnovate, a leading company in the technology sector, launched a training programme for employees on AI last year. According to the CEO of TechInnovate, ‘we see AI as a crucial factor in our future strategy.’ [1] Additionally, the company started a pilot project with AI agents in June 2025, which is a significant step towards full implementation. [1]
Guidelines for Safe Implementation
The Dutch government issued guidelines in March 2025 for the safe implementation of AI in businesses. These guidelines provide clear instructions for the use of AI agents, focusing on data privacy, ethical use, and technical knowledge. [1] Jan Jansen, CEO of LEGALFLY, emphasises that ‘successfully implementing AI agents requires thorough preparation and a good understanding of the technology.’ [2] LEGALFLY has therefore developed a set of instructions to help companies integrate AI agents, officially launched at a conference in Amsterdam on 15 July 2025. [2]
Specific Applications in Various Sectors
AI agents are already being successfully deployed in various sectors. Zoho, a well-known software company, launched Zia LLM, its own large language model, and Zia Agents, a platform with ready-to-use AI agents for different business functions. For example, Zia Agents offers the Revenue Growth Specialist, which identifies opportunities for upselling and cross-selling, and the Deal Analyzer, which analyses sales deals and suggests follow-up actions. [3] AWS also introduced a new AI platform, Amazon Bedrock AgentCore, which helps companies develop and manage agents. [4] In the transportation sector, 61% of organisations expect fully autonomous agents to operate independently to achieve specific goals within the next five years. [5]
Benefits and Challenges
The integration of AI agents offers both benefits and challenges. Benefits include improved efficiency, reduced errors, and increased availability of personalised services. LEGALFLY, for instance, provides AI agents that can review, redline, and turn around legal documents in minutes, and automatically flag risks. [2] However, challenges such as skill shortages, integration issues, and data quality are cited as obstacles. In the transportation sector, 49% of organisations identify skill shortages as a major problem. [5] Privacy also plays a crucial role. LEGALFLY ensures that client data is anonymised before processing to prevent any confidential information from being shared with LLMs. [2]
Practical Examples and Future Perspectives
Practical examples illustrate the potential of AI agents. For instance, Google attracted Varun Mohan, founder and head of AI startup Windsurf, for $2.4 billion. Windsurf is known for ‘vibe coding’, a new programming method using AI tools. [6] OpenAI launched a new AI agent for ChatGPT on 17 July 2025, capable of performing complex tasks, including in-depth financial research and filling out online forms. [7] In the aviation sector, Delta Airlines began using AI on 17 July 2025 to determine individual ticket prices, yielding remarkable results. [6] Future plans include regular improvements and expanding the application of AI agents to more sectors and user groups. [7]