Goldman Sachs interprets Meta (META.US) AI paradigm transformation, MuseAI opens a new era of intelligent agents

The Zhitong Finance App learned that Meta (META.US) recently launched the personal AI agent product Muse AI, marking the official leap of consumer artificial intelligence from question-and-answer chatbots to the era of intelligent devices that can independently complete realistic tasks. Goldman Sachs released a research report to maintain Meta's “buy” rating. The 12-month target price is $725, which has room to rise 11.9% from the stock price of $648.03 at the time of the report, while also analyzing product potential, competitive barriers, profit prospects, and potential risks.

“Dialogue to Action”: Muse AI Reshapes C-Side AI Product Form

Muse AI is different from traditional conversational robots. Relying on the Muse Spark 1.3 model, it can independently browse websites, fill out forms, complete purchases, and implement a new “dialogue to action” paradigm, covering high-value personal workflows such as travel reservations, billing negotiations, and administrative matters. The product opens up multiple apps, websites, and WhatsApp, and uses message interaction logic to lower the user's entry threshold.

At the commercial level, Meta launched the C-side AI tiered subscription model for the first time. It has three levels: free, $20 per month powerful edition, and $100 per month premium edition. It is expected to expand monetization channels such as smart e-commerce transaction sharing and advertising in the long term. In terms of security mechanisms, the product uses isolated virtual machine operation tasks, combined with the Sentinel permission control system, and uses WhatsApp encryption technology to isolate data and advertising systems from each other. However, consumers' privacy concerns are still a core obstacle to large-scale popularization of products.

Ecological distribution builds a moat, and the industry pattern ushered in platform-based concentration

Meta's greatest strength comes from distribution barriers brought about by the social ecosystem. Relying on the billions of active Facebook, Instagram, and WhatsApp users, it integrates the complete infrastructure of accounts, payments, and advertising to form a competitive advantage that independent AI startups cannot catch up with. According to the research report, Meta will prioritize expanding the number of users at this stage and will not rush to pursue commercialization benefits.

At the industry level, the competitive focus of consumer AI has moved away from simple model performance competition, and the ability to integrate platform channels and business scenarios is becoming more and more critical. Referring to the mobile Internet iOS and Android pattern, in the future, the AI intelligent agent market may be concentrated on leading platforms, and Meta and Google have the innate conditions for large-scale development. At the same time, the industry still has problems to be solved: existing web pages and applications have not yet adapted to AI agents, browser simulation and API calls, and the industry has yet to reach a unified consensus.

Heavy investment in AI suppresses short-term financial reports, looking at long-term performance recovery

Meta is shouldering huge capital expenses for the AI business. Goldman Sachs predicts that the company's revenue will grow from $200,965 billion in 2025 to $37,837 billion in 2028. However, investment in computing power infrastructure will suppress profits, and free cash flow may turn negative in 2027, and is not expected to be repaired until 2028; with the release of scale effects, EBITDA will experience accelerated growth, and valuations will gradually be digested as performance increases.

Meta still faces many downside risks: competition on the Internet circuit intensifies; antitrust and regulatory scrutiny may limit merger and acquisition expansion; there is a risk that businesses such as metaverse and social e-commerce will fall short of expectations; and if AI capital expenditure suppresses profit margins for longer than market expectations, it will also damage market confidence.

The September Connect developer conference and the release of the next-generation basic model will be an important observation window. Overall, Muse AI is Meta's key layout to open up AI and consumer scenarios, but at this stage, the focus is on strategic cards. Commercialization will take time, and investors need to balance the operating pressure brought about by AI technology dividends and high investment.