Meta has announced significant updates for its Muse AI agent, making it more capable and pervasive, while California addresses the growing environmental impact of data centers, essential infrastructure for the expansion of artificial intelligence.
What happened
During the Meta Connect 2026 event, the company unveiled a significant evolution of its Muse AI agent. This virtual assistant, already known for its ability to support users with shopping, will now be able to manage dedicated email addresses and interact via video chat, greatly expanding its scope Meta is making Muse more powerful and will let you video chat with it, too. The goal is to integrate Muse more deeply into users' daily lives, transforming it into a true "agent" capable of performing complex and proactive tasks Meta Connect 2026: The biggest news and announcements. This expansion of capabilities raises questions about user autonomy and control over personal finances, as highlighted by early experiences describing Muse as "great at spending my money" Meta’s AI agent is a cute little guy who’s great at spending my money.
In parallel with this innovation in AI agents, another news item underscores the growing infrastructural challenges related to artificial intelligence. California Governor Gavin Newsom signed a series of bills aimed at mandating greater transparency regarding the impact of data centers in terms of electricity and water consumption Data centers are black boxes, but California wants to change that. These facilities, crucial for powering generative AI and complex models like those underlying Muse, are proliferating and generating protests from local communities concerned about the depletion of water resources and rising energy bills. The Californian legislation aims to give local communities more say and access to data on the environmental impact of these technological "black boxes."
Why it matters
The advancement of AI agents like Muse represents a significant step towards a more fluid and autonomous human-machine interaction. While it promises to simplify the management of many daily activities, it also raises crucial questions about privacy, data security, and individuals' decision-making autonomy. An agent capable of managing emails and purchases, if not adequately governed, could erode personal control and increase vulnerability to manipulation or errors. Dependence on such systems requires deep reflection on control mechanisms and accountability in the event of incorrect decisions or those not aligned with user interests.
The environmental impact of data centers, on the other hand, is a sustainability issue that can no longer be ignored. The training and operation of increasingly large AI models require colossal amounts of energy and water. California's decision to demand greater transparency is an important signal: the growth of AI cannot come at the expense of natural resources and community well-being. The lack of public data on the consumption of these infrastructures has so far hindered accurate assessment and sustainable planning. The expansion of AI, if not accompanied by robust AI governance and sustainable practices, risks creating a gap between technological benefits and environmental and social costs.
The HDAI perspective
These two seemingly disparate developments converge on a fundamental point for Human Driven AI: the urgency of an ethical and responsible approach to artificial intelligence. The evolution of agents like Muse pushes us to reflect on how to design systems that augment human capabilities without replacing critical judgment or compromising individual autonomy. It is essential that the design of these agents includes mechanisms for transparency, user control, and accountability, ensuring that AI serves people and not the other way around.
At the same time, the issue of data centers reminds us that AI is not an abstract entity but has a tangible physical and environmental footprint. Technological innovation must be intrinsically linked to sustainability. The demand for transparency on the energy and water consumption of data centers is a crucial step towards greater corporate responsibility and infrastructural planning that considers the needs of communities and the planet. Ethical AI is not just about algorithms; it's about the entire ecosystem that supports it, from agent design to infrastructure sustainability. These themes will be central to discussions at the HDAI Summit 2026, where we will explore how to balance technological progress and social responsibility.
What to watch
It will be crucial to observe how Meta continues to develop Muse and how users react to an increasingly autonomous AI agent. Similarly, the implementation of new laws in California will provide a model for other regions facing similar challenges related to the environmental impact of AI infrastructure.