The EU is banning the use of AI-powered virtual assistants during online meetings. Such assistants are often used to transcribe, take notes, or even record visuals and audio during a video conference.
In a presentation from the European Commission delivered to European Digital Innovation Hubs earlier this month, there is a note on the “Online Meeting Etiquette” slide that states “No AI Agents are allowed.”
AI agents are tools that can perform complex, multi-step tasks autonomously often by interacting with applications, such as video conferencing software. For example, Salesforce uses AI agents to call sales leads.
The Commission confirmed this presentation was the first time this rule had been imposed but declined to explain why when questioned by Politico. There is no specific EU legislation that covers AI agents, but the AI models that power them will need to abide by the strict and controversial rules of the AI Act.
AI agents raise security concerns
While AI notetakers and other agent types are not inherently a security threat, according to a 2025 report from global AI experts, security risks stem from the user being unaware of what their AI agents are doing, their innate ability to operate outside of the user’s control, and potential AI-to-AI interactions. These factors make AI agents less predictable than standard models.
SEE: How Can AI Be Used Safely? Researchers From Harvard, MIT, IBM & Microsoft Weigh In
Tech companies do have to be cautious when promoting products that can accomplish an increasing amount without the user’s awareness. One of the biggest cautionary tales is that of Microsoft Recall, an AI tool that allowed users to control their PC or search through files using natural language.
The convenience came at a cost: Recall captured screenshots of active windows every few seconds, saving them as a timeline, raising concerns about privacy and data usage and leading to significant launch delays. Microsoft has since released a series of agents specifically designed to tackle cyber threats.
AI agents are growing in prevalence
This hasn’t stopped the AI players from handing over more control to their models. Anthropic added a Computer Use feature to its Claude Sonnet chatbot in October 2024, which gave it the ability to navigate desktop apps, move cursors, click buttons, and type text. Its deep research function, announced this week, also responds to prompts “agentically,” as does Microsoft’s equivalent.
Last month, OpenAI expanded its text-to-speech and speech-to-text tools to agentic models, indicating their growing relevancy. In January 2025, OpenAI announced Operator, an agentic tool that runs in-browser to autonomously perform actions such as ordering groceries or booking tours.
SEE: EU Invests €1.3 Billion to Boost AI Adoption & Improve ‘Digital Competencies’
Anthropic and OpenAI are even working together to improve agent technology, with the latter adding support for the former’s Model Context Protocol, an open-source standard for connecting AI apps, including agents, to data repositories. Anthropic has also joined forces with Databricks to help large corporate clients build their own agents.
TechRepublic predicted at the end of 2024 that the use of AI agents will surge this year. OpenAI CEO Sam Altman echoed this in a January blog post, saying “we may see the first AI agents ‘join the workforce’ and materially change the output of companies.”
By 2028, 33% of enterprise software applications will include agentic AI, up from less than 1% in 2024, according to Gartner. A fifth of online store interactions and at least 15% of day-to-day work decisions will be conducted by agents by that year.
