Bank of America Backs Meta Stock, Sets $810 Target on Muse Launch

Bank of America reiterated its Buy rating on Meta Platforms (META) and set an $810 price target, betting that Muse, the company's new consumer AI agent, can reprice the story around its multibillion-dollar AI infrastructure spending.

The $810 target represents 9.3% upside from BofA's $741.25 reference price, though the potential return narrows to roughly 8% using the Sept. 22 intraday price. In a note shared with The Street, BofA said its new Muse offering "might finally put to rest many of the concerns surrounding Meta" if it can deliver a differentiated consumer product.

The early numbers

Meta launched Muse on Sept. 8. The assistant can browse websites, organize email, and complete multi-step tasks with limited user intervention, pushing the company beyond chatbots toward a practical digital agent.

Citing Sensor Tower, BofA said Muse surpassed 2.5 million cumulative U.S. downloads by Sept. 19 and reached approximately 557,000 daily active users. Its 1.43 million iOS downloads through day 12 also edged ChatGPT's 1.37 million at the same stage.

Investors have taken notice. Meta stock has risen 36% over the past month, dwarfing the broader market's gain of 1.2%, according to Seeking Alpha data.

Muse's ability to execute multistep tasks — managing email, booking travel, completing transactions — separates it from conventional chatbots. Meta can eventually promote the agent across its platforms, which serve 3.6 billion daily users, creating a distribution advantage few competitors can match. Independent testing found Muse genuinely useful but potentially uncomfortable due to the personal access it requires.

Robust retention could lift Meta's valuation before Muse contributes meaningful revenue. Downloads alone, however, will not prove lasting demand.

A multibillion-dollar compute decision

Muse also forces a strategic question: Should Meta use its limited AI computing power for its own products, or sell that capacity to outside customers?

BofA estimates that external leasing could generate up to $40 billion per gigawatt, based on recent industry transactions. Leasing would create visible revenue and provide faster evidence that Meta's infrastructure buildout can earn an acceptable return. Management, however, believes consumer agents and AI-enhanced advertising could ultimately create greater value.

The stakes are massive. Meta expects 2026 capital expenditures of $130 billion to $145 billion, compared with $72.22 billion during 2025, Reuters noted. Every unit of compute is therefore an opportunity-cost decision, not just a technical allocation.

Muse currently offers a free tier alongside $20 and $100 monthly subscriptions. BofA does not expect subscriptions, advertising, or commerce commissions to become financially meaningful before 2028. Jefferies framed the longer-term opportunity differently: its scenario assumes one billion users by the end of 2027 and 3% paid conversion, producing $10.8 billion in annualized revenue.

Prioritizing Muse, therefore, sacrifices potentially lucrative leasing income for a broader platform bet spanning subscriptions, commerce, APIs, and advertising. Early engagement will help determine whether that trade pays off.

Valuation leaves little room for error

Meta stock no longer carries the deep discount that protected investors earlier in the year. BofA said shares trade near 21 times its 2027 earnings estimate, up from approximately 16 times several months ago. The $810 target applies a 24-times multiple to estimated 2027 GAAP earnings, a slight premium to the S&P 500. With only about 8% upside from the intraday level on Tuesday, Sept. 22, the margin for disappointment has narrowed considerably.

That makes Muse a supporting catalyst rather than a complete investment thesis. Investors need to see downloads translate into durable usage, paid subscriptions, commerce activity, or measurable advertising gains before pricing in further multiple expansion.

The risks are interconnected. Muse requires access to sensitive accounts, creating privacy and trust concerns. Amazon has already blocked the agent from shopping on its platform, demonstrating how third-party restrictions could weaken Muse's usefulness and commerce ambitions. Meanwhile, Meta's $130 billion to $145 billion spending plan pressures free cash flow and raises the cost of misjudging AI demand.

Investors should monitor retention, paid conversion, task-processing costs, partner access, and 2027 spending guidance. Muse has improved Meta's AI narrative. At this valuation, it must become a monetizable habit instead of just an impressive demonstration.