There has been much debate over Meta Platforms' (NASDAQ: META) artificial intelligence (AI) strategy. The Magnificent Seven company’s AI capabilities are clearly benefiting its advertising business, with revenue growth accelerating throughout 2025. However, Meta’s LLaMa models have generally not impressed tech experts. This, along with Meta’s steep Reality Labs losses, raises questions about how well Meta can monetize AI long-term.
However, Meta just released a new AI model: the Segment Anything Model 3 (SAM 3). SAM 3 is turning heads in the tech community, but few investors are talking about it. Analysts at Citizens JPM are one exception. On Nov. 24, the firm issued a $900 price target on Meta, supported by its assessment of SAM 3. Notably, Citizens' price target implies 36% upside in Meta shares versus the Dec. 4 close.
SAM 3 represents a meaningful advance in Meta’s AI capabilities, with potential applications across computer vision, content creation, and automation. Its release highlights Meta’s continued investment in foundational AI research and reinforces the company’s position as a competitive player in the evolving AI landscape.
SAM 3: Meta’s Data Annotation Workhorse
SAM 3 is a computer vision model. It allows users to detect, segment, and track objects in images and videos using simple text prompts. The company’s Segment Anything Playground allows anyone to see the model in action. It appears that Meta now has a top-tier computer vision model. SAM 3 currently holds the top spot on Roboflow’s AI Vision Model Rankings board. AI analyst Nate B. Jones calls SAM 3 a “ChatGPT moment” for video and other applications. It may not be immediately clear why this is so important, but SAM 3’s utility sits at the core of AI development: training models.
Segmenting and tracking objects in images and videos is vital to generating data that can train multimodal AI models. Researchers have often generated this data by employing humans to annotate images and videos manually. SAM 3 makes this process much faster while approaching human-like accuracy.
Citizens cites this increased annotation speed as a key reason for its $900 price target. Faster annotation can greatly increase the amount of data Meta has to train its other systems.
Short-form videos have become highly prevalent on Instagram and Facebook. Using SAM 3, Meta can analyze and generate data from those videos at scale. It can then improve its content and advertising recommendation models by training them with that data, ultimately boosting engagement on Facebook and Instagram, making users more likely to interact with advertisements. All this goes back to increasing the value of advertising on Meta’s apps, driving higher revenue for the firm.
The company could also utilize the data generated by SAM 3 to enhance its general-purpose LLaMA models.
SAM 3 Could Drive Value Everywhere from Instagram to Ray-Ban
Aside from generating data to train other models, SAM 3 has many practical implications that can benefit Meta. Because it can detect visual objects so well, SAM 3 serves as a tool for image/video editing and creation.
Meta notes that SAM 3’s capabilities will soon be available in Edits, Instagram’s video creation app, helping content creators spend less time editing. By enabling faster content creation, Meta reinforces the powerful network effects on its apps. The more content people create and share on a platform, the more likely users are to find content they engage with. As users engage more, they incentivize creators to make more content. This positive feedback loop is a huge advantage for Meta, keeping both users and creators on its platform.
Meta also provides AI tools that marketers can use to generate visual advertisements. It calls these tools its Advantage+ Creative suite. Meta could potentially use SAM 3 to improve the capabilities of Advantage+ Creative, which could make advertisements more engaging and targeted, increasing the value of advertising on Meta’s apps.
Lastly, SAM 3 can likely improve the ability of Meta’s AI glasses to create augmented reality overlays or quickly detect a lost item in a room.
Meta’s Displays AI Chops With SAM 3
Overall, it is important to consider the progress that Meta has achieved through SAM 3. It shows that the firm can make models that technologists believe rival the best in their domain. This should give investors confidence in Meta’s AI strategy going forward.
Additionally, SAM 3 has the potential to significantly improve every area of Meta’s business, from advertising to AI glasses. This is huge, showing the synergistic nature of Meta’s AI investments.
Where Should You Invest $1,000 Right Now?
Before you make your next trade, you'll want to hear this.
MarketBeat keeps track of Wall Street's top-rated and best performing research analysts and the stocks they recommend to their clients on a daily basis.
Our team has identified the five stocks that top analysts are quietly whispering to their clients to buy now before the broader market catches on... and none of the big name stocks were on the list.
They believe these five stocks are the five best companies for investors to buy now...
The article "Meta’s AI Moment? New SAM 3 Model Has Wall Street Turning Bullish" first appeared on MarketBeat.
