Bank of America Keeps Meta Buy Rating, $810 Target on AI Push

Bank of America reiterated its Buy rating on Meta (META) with an unchanged $810 price target, implying roughly 20% upside from the stock's September 16 close of $673.31, according to a note shared with TheStreet.

The call follows two moves Meta made last week that Wall Street has been watching for: one aimed at cutting the cost of running AI, the other at charging users for it. Together they give investors the most concrete answer yet to the question hanging over the stock all year — when Meta's hyperscale spending starts paying back.

The chip strategy

Meta confirmed it will deploy its third-generation custom AI chips, called Arke, in the first half of 2027. A fourth-generation chip, Astrid, follows in late 2027. Both were built with Broadcom for AI inference — the process where trained models generate responses — Bloomberg reported.

Meta's vice president of engineering told Bloomberg the chips are engineered to outperform "whatever Nvidia is currently shipping" on a per-watt and per-dollar basis. Meta plans more than one gigawatt of custom chip capacity over the next 12 months.

The company also tried to build a chip that could handle both training and inference. It canceled that project after finding the design would cost roughly 30% more.

Bank of America ran the numbers on the savings. If Meta deploys five to six gigawatts of owned capacity in 2027 at $200 billion in total infrastructure spend, with chips making up 60% of that cost, custom silicon at 40% savings versus third-party alternatives could mean roughly $8.5 billion saved. That is a model estimate, not company guidance.

Google's TPU program offers a precedent. Custom silicon took years to pay off for Alphabet but now anchors its AI infrastructure cost management. The Arke and Astrid chips arriving in 2027 are the first real test of whether the economics hold for Meta too.

Broadcom CEO Hock Tan put it bluntly on his earnings call. "When you co-develop a chip that is optimized for your particular LLM workloads, you will outperform any GPU," he said, adding that customers can do it "at half the cost." He confirmed Meta's program is on track and that three MTIA generations will ship by end of 2027.

Meta One brings a first revenue signal

Meta launched Meta One on September 15, a global subscription service bundling AI features, customization tools and creator capabilities across Instagram, Facebook, WhatsApp and Meta AI. More than 50 features shipped with it. The service has already recorded 15 million subscriptions and trials.

Consumer plans start at $2.99 per month for individual apps and reach $19.99 for the Premium bundle. Creator and business plans run from $14.99 to $499 per month.

The 15 million figure mixes trials and paying subscribers, and nobody knows yet how many convert to recurring revenue. But it is the first signal that users will pay for features on platforms they have used for free.

Bank of America's math is straightforward. Every 1% of Meta's 3.6 billion users subscribing at $10 per month in average revenue adds roughly $4.3 billion a year — about 1.2% upside to 2028 revenue estimates. Snapchat Plus has reached 5.5% daily active user penetration. That is the benchmark Meta is chasing.

Two catalysts remain

Bank of America flagged five catalysts for Meta at the start of its coverage. Three are now announced: Muse, custom chips and Meta One.

Two remain. The Connect conference comes first. No major AI announcements are expected there, though any update on Muse adoption would matter. Muse is Meta's AI image and video generation system, and how quickly it pulls users into paid tiers has drawn less attention than it deserves.

The bigger catalyst is Watermelon, Meta's upcoming frontier LLM. A strong model could improve Meta's advertising systems and open API licensing fees as a new revenue line. It would also be the clearest signal yet that Meta's AI research produces financial outcomes, not just benchmark scores.

The risks

The chip savings are a model. Meta One conversion is unknown. Watermelon has not shipped. All three are bets, not bookings.

Meta still earns the vast majority of its revenue from digital advertising. A macro downturn or a pullback in ad spending hits the whole thesis. The company's fixed asset base has also grown fast enough that cost flexibility in a downturn is more limited than it used to be.

A competition risk sits beneath the advertising risk. AI-native platforms are starting to compete for the same user attention Meta's apps have dominated for years. If engagement shifts, the ad revenue picture changes before any of the new revenue lines have scaled enough to compensate.

The $810 target equals 24 times the 2027 GAAP earnings estimate — a premium to the broader market. Whether Meta earns it depends on whether the chip economics land, the subscription business scales and Watermelon actually performs. None of that is settled yet.