The race to develop AI agents capable of performing tasks autonomously is intensifying among major technology companies, with various approaches emerging amid uncertainty over which model will prevail. These AI agents, designed to act on users’ behalf to manage routine or complex activities, are seen by industry leaders as a potentially transformative gateway to the digital world and a source of significant commercial influence.
Several tech giants have introduced or enhanced AI agents in recent months. Meta launched Muse, a digital assistant focused on managing everyday tasks, while OpenAI introduced dots targeting more sophisticated work assignments. Microsoft revamped its Copilot assistant to operate more independently, Apple incorporated agentic features into its Siri assistant with its latest iOS update, and Google released a new productivity-focused assistant as part of its Gemini project. In addition, startups such as Instinct have begun testing AI agents on an invitation-only basis.
The contrast between Muse and dots highlights differing strategic visions. Meta positions Muse as a general-purpose helper designed to facilitate daily activities and is looking to expand its role in ecommerce by enabling transactions that generate revenue through commission. Analysts at Jefferies note that Muse currently interacts with two-thirds of travel, commerce, and reservation sites, reflecting Meta’s effort to link users more directly with merchants in a bid to influence buying behavior on a large scale.
OpenAI, by comparison, is concentrating on high-end professional use cases, with CEO Sam Altman emphasizing tasks typically assigned to “high-agency engineers or chiefs of staff.” The company plans to monetize this approach through premium subscriptions, with dots available initially only to ChatGPT subscribers paying at least $100 per month.
Other companies’ offerings tend to fall between these poles. Microsoft’s Copilot now offers users options including an Autopilot mode that can handle extended work tasks through multiple tabs, illustrating an effort to balance user autonomy with guided interaction. Muse adopts a more proactive strategy by frequently suggesting tasks, while OpenAI’s dots present a minimal interface for users to explore capabilities independently.
A key challenge across all players is encouraging widespread adoption of these AI agents, which will require users to alter longstanding digital habits and develop trust in the technology. Previous attempts to introduce AI features that monitor user activity have sparked privacy concerns, underscoring the need for cautious product design.
The market outlook also raises questions about whether consumers will prefer a single, adaptable AI agent to manage all their needs or a collection of specialized agents embedded within different applications. Companies appear to be hedging their bets: Walmart operates its own shopping assistant, Sparky, while partnering with Meta to integrate with Muse. Conversely, Amazon currently restricts access to its platform for agents developed by Anthropic, OpenAI, and Meta.
Should one or more companies succeed in establishing a dominant AI agent ecosystem, they could achieve significant market power comparable to existing leaders in search, social media, and ecommerce. However, the ultimate outcome remains uncertain as firms compete to define the future interface between users and artificial intelligence.