Some links on this page are affiliate links: if you buy through them we may earn a commission, at no extra cost to you.
Meta did not fully acquire Scale AI. It made a minority investment of roughly $14 billion—initially reported as $14.8 billion and later widely described as $14.3 billion—for 49% of the artificial-intelligence data company. Scale AI’s founder, Alexandr Wang, also moved to Meta to work on its expanding superintelligence effort.
The transaction combined three bets: access to sophisticated AI data and evaluation capabilities, recruitment of a prominent founder, and a faster organizational push to compete with OpenAI, Google and other frontier-AI companies.
The deal in brief
- Investor: Meta
- Company: Scale AI
- Stake: 49%, according to reporting
- Amount: Approximately $14.8 billion in initial reports; later reporting described the finalized investment as $14.3 billion
- Valuation: More than $29 billion, according to Scale’s announcement
- Structure: Minority investment, not a full acquisition
- Leadership: Scale co-founder Alexandr Wang joined Meta; Jason Droege became Scale’s interim CEO
- Announcement: Scale confirmed the transaction on June 12, 2025
The most accurate description is that Meta invested roughly $14 billion for a major minority stake in Scale AI while recruiting Wang. Calling it a $14.8 billion acquisition overstates both the final reported amount and Meta’s ownership.
Recommended Free Tools
Scale said it would remain an independent company, continue serving its customers and keep its operations separate from Meta.
#1 Best Overall
- Built for Local AI Development: AMD Ryzen AI Halo is designed for local AI development and inference, featuring 128GB unified memory and support for up to 200B parameter models to build and run intensive AI workloads locally.
- 128GB Unified Memory: Features 128GB LPDDR5x unified memory at 8000 MT/s with 256 GB/s memory bandwidth, providing a shared memory pool across the CPU, GPU, and NPU to support larger AI models.
- AMD Ryzen AI Max+ 395 Processor: Features 16 cores, 32 threads, and Zen 5 architecture, paired with AMD Radeon 8060S integrated graphics featuring 40 RDNA 3.5 compute units and an AMD XDNA 2 NPU with up to 50 TOPS.
- Linux AI Developer Platform: Purpose-built for Linux-based AI development with full AMD ROCm software support and preloaded tools, models, and workflows optimized for local AI development.
- Compact, Connected Design: Includes a 2TB M.2 SSD, 10GbE LAN, Wi-Fi 7, Bluetooth 5.4, USB-C connectivity, and HDMI 2.1b.
Why the numbers changed from $14.8 billion to $14.3 billion
On June 10, 2025, initial reports said Meta had agreed to invest approximately $14.8 billion for 49% of Scale AI. Two days later, Scale officially confirmed a significant new investment from Meta, but did not state the $14.8 billion figure in its announcement.
Later coverage generally described the finalized investment as $14.3 billion. Scale said the transaction valued the company at more than $29 billion. These figures are not necessarily contradictory: the first number came from early reporting, while the later figure reflected how the completed transaction was subsequently described.
For readers assessing Meta’s financial commitment, the sensible shorthand is roughly $14 billion for 49% of Scale AI, with $14.8 billion identified as the initial reported figure and $14.3 billion as the later widely reported amount.
Crashes, No Sound, or Screen Glitches?
Random freezes, missing sound and display glitches usually trace back to one bad driver. Find and replace yours safely.Free scan · under a minutePC Slower Than It Used to Be?
A free scan shows the junk files, broken settings and background clutter dragging Windows down - then fixes them in one click.Free scan · Windows 10 & 11What Scale AI actually does
Scale AI is more than a conventional data-labeling company. It helps organizations collect, prepare, annotate and evaluate the data used to develop artificial-intelligence systems.
Its work can include:
- Labeling text, images, video, audio and other multimodal data.
- Preparing domain-specific datasets for model training.
- Collecting human feedback to improve model responses.
- Evaluating whether models are accurate, reliable and useful.
- Red-teaming systems to identify safety and security weaknesses.
- Supporting enterprise, robotics, autonomous-vehicle and government AI projects.
- Helping customers move AI systems from experimentation toward deployment.
Scale’s product and services portfolio includes Scale Data Engine, Scale GenAI Platform and Scale Donovan. The company’s relevance comes from its position across the AI development lifecycle: data preparation, training support, testing, evaluation and deployment.
That matters because a powerful model is not created solely by adding more computing power. The quality, diversity and accuracy of the data—and the methods used to test a model—can influence how well it performs in the real world.
Why Meta would invest so much
1. Better data and evaluation
Meta develops large AI models and consumer AI products. A strategic relationship with Scale could help it obtain or develop higher-quality, specialized and human-verified data for training, evaluation, safety testing and product improvement.
That does not mean Meta automatically obtained all of Scale’s datasets or unrestricted access to customer projects. Scale said Meta would not receive access to its internal systems or other customers’ confidential information. The potential benefit is better access to expertise, services and a strategically important supplier—not ownership of every piece of data Scale handles.
2. Alexandr Wang’s talent and recruiting value
Wang co-founded Scale AI and had served as its chief executive. His move gave Meta a founder-operator with experience building an AI-focused company, working with major customers and recruiting technical talent.
Contemporary reports described Wang as potentially playing a major role in Meta’s new superintelligence organization. The precise scope of his authority should not be overstated, however. The available evidence supports saying that he joined Meta’s AI efforts, not that he was formally announced as the head of all Meta AI.
3. A catch-up and urgency strategy
Meta had substantial AI infrastructure, research talent and distribution through Facebook, Instagram, WhatsApp, Messenger, Threads and its hardware products. It nevertheless faced pressure to move more decisively as OpenAI, Google and other companies competed for leadership in frontier models and AI assistants.
The Scale investment fit into a wider effort by Mark Zuckerberg to recruit talent, reorganize AI teams and make Meta’s AI ambitions more urgent. The transaction was therefore not simply a data purchase. It was also an organizational and competitive statement.
4. A deeper commercial relationship
Scale said the transaction would substantially expand its commercial relationship with Meta. One later report said the agreement required Meta to spend at least $500 million annually on Scale data for five years. That figure came from a source familiar with negotiations rather than a publicly disclosed contract, so it should be treated as reported—not as an officially confirmed term.
A larger commercial relationship could give Scale a valuable anchor customer while giving Meta a closer connection to an important AI supplier. It could also increase questions about whether Scale remains sufficiently neutral for Meta’s competitors.
Rank #2
- EVOLUTION AMD RYZEN AI MAX+ 395 MINI PC - GMKtec EVO-X2 is the next evolution in AI mini PC Ryzen Strix Halo series. Thanks to AMD Simultaneous Multithreading (SMT) the core-count is effectively doubled, to 32 threads. Ryzen AI Max+ 395 has 64 MB of L3 cache and can boost up to 5.1 GHz, depending on the workload. The Ryzen AI Max+ 395 is currently rated as the "most powerful x86 APU" on the market for AI computing.
- AI NPU with XDNA 2 ARCHITECTURE - Powered by 16 “Zen 5” CPU cores, 50+ peak AI TOPS XDNA 2 NPU and a truly massive integrated GPU driven by 40 AMD RDNA 3.5 CUs, the Ryzen AI MAX+ 395 is a transformative upgrade and delivers a significant performance boost over the competition. The Ryzen AI Max+ 395 excels in consumer AI workloads like the llama.cpp-powered application: LM Studio. Shaping up to be the must-have app for client LLM workloads, LM Studio allows users to locally run the latest language model without any technical knowledge required and unleash their creativity and productivity.
- AMD RADEON 8090S iGPU GAMING PC - The AMD Radeon RX 8060S offers all 40 CUs with up to 2.9 GHz graphics clock and uses the new RDNA 3.5 architecture. The powerful iGPU is positioned between an RTX 4060 and 4070 laptop GPU and therefore enables gaming in FHD at maximum details in most demanding games. The 8060S can also utilize the full 64GB pool, which is perfect for running LLMs such as Deepseek 32B, which runs comfortably on this machine.
- EIGHT CHANNEL LPDDR5X - LPDDR5X is a new ground breaking memory small form factor installed on-board. With blazing speeds up to to 8000MT/s, it runs 1.5x faster than the DDR5 SODIMMs; 90% better performance over DDR5 SODIMMs in video conferencing and photo editing; 30% better performance in productivity apps; 4% better performance in digital content workloads.
- QUAD SCREEN 8K DISPLAY SUPPORT - EVO-X2 AI Mini PC support 4-screen 4K/8K output via HDMI 2.1 (8K@60Hz), DisplayPort 1.4 (4K@60Hz), and dual USB 4 40Gbps Transfer speed (supporting PD3.0/DP1.4/DATA). Ideal for gaming, video editing, and multitasking, it provides expansive and crisp multi-display support.
5. Strategic influence without full ownership
A 49% stake allows Meta to make an unusually large economic commitment without formally integrating Scale into Meta. This structure may preserve Scale’s existing business and operating model while giving Meta substantial strategic and financial exposure.
The Tool Desk
Outbyte PC Repair FREERepair Windows errors before they cause bigger problemsFix Now →Outbyte Driver Updater FREEScan for outdated or missing drivers - takes under a minuteDriver Scan →That distinction is important. A minority investment can create influence, commercial dependence or governance concerns even when the invested company remains legally and operationally independent.
What “superintelligence” meant in 2025
In general AI discussions, superintelligence refers to a hypothetical system whose capabilities exceed those of humans across many important intellectual tasks. In June 2025, Meta’s use of the term described an ambition and an organizational goal—not a demonstrated achievement.
It should not be read as proof that Meta had built human-level artificial general intelligence or a superhuman system. The Scale investment did not itself create superintelligence.
Meta later formalized this effort as Meta Superintelligence Labs. In 2026, Meta said the organization had produced Muse Spark, described as the first model in a new series, and connected the project to its goal of “personal superintelligence.”
Do these 3 things before closing this tab:
1Scan for outdated or missing drivers - takes under a minute2Repair Windows errors before they cause bigger problems3Fix the driver behind crashes, sound loss and screen glitchesMeta’s later framing emphasizes AI assistants that can reason, use tools and act on a user’s behalf across Meta’s apps and devices. Its announcements about Muse Spark, agentic features and expanded computing infrastructure show organizational follow-through, but they do not prove that the Scale investment alone caused those results.
Relevant updates include Meta’s announcements about Muse Spark and Meta Superintelligence Labs, agentic Meta AI capabilities and its AI infrastructure strategy.
What happened to Scale AI and Alexandr Wang?
Wang left the Scale CEO role to join Meta’s AI efforts. Jason Droege became interim CEO, while Wang remained on Scale’s board. Scale said it would continue operating independently and serving AI labs, enterprises and government customers.
This arrangement creates a practical governance question: Wang could have responsibilities at Meta while retaining a board role at Scale. That fact alone does not establish wrongdoing, but it makes questions about confidentiality, conflicts and customer confidence more important.
Scale also said Meta would not receive access to other customers’ confidential information and that its operations would not be integrated with Meta. Those safeguards address direct data access, but they do not eliminate every concern about commercial influence or customer perception.
The biggest risk: customer trust
Scale served customers across the AI industry, including companies that compete with Meta. After Meta’s investment, customers could reasonably ask whether Scale could remain neutral, even if contractual and technical protections were maintained.
Reports said Google and OpenAI were considering reducing or ending relationships with Scale. Those reports should not be generalized into a confirmed mass departure, and commercial decisions can change over time. The underlying risk is nevertheless clear: a vendor that works with competing AI companies may become less attractive when one competitor takes a large stake and recruits its founder.
Scale’s independence has at least four dimensions:
Free tools Windows power users keep installed
One-click scans. No signup required.
- Legal independence: Scale was not fully acquired by Meta.
- Operational independence: Scale said it would continue running separately.
- Data separation: Scale said Meta would not receive other customers’ confidential information.
- Perceived neutrality: Customers may still question whether Meta’s investment influences priorities, access or long-term strategy.
The fourth issue can be as important as the first three. In a market where data and evaluation providers work with several competing model developers, trust is part of the product.
Rank #3
- Intel Core Ultra 9 285 Processor: Newly developed cores deliver ultra-smooth and responsive gameplay. AI accelerators prepare users for the next era of gaming on an AI PC.
- Simplistic Design: Enjoy the latest generation of Windows 11 Home for your everyday needs. *MSI recommends Windows 11 Pro for business use.
- NVIDIA GeForce RTX 5070 Ti GPU
- Cool While Gaming: In conjunction with an RGB CPU Air Cooler, the Aegis RS features four system cooling fans; three in the front and one in the rear to pull in cool air and push heat out of the PC.
- Turn on the Bright Lights: With the built-in RGB lighting, take your gaming experience to the next level by pressing the MSI LED button to cycle through lighting options. Customize lighting even further with MSI Center software.
Regulatory and execution risks
Regulatory scrutiny
A multibillion-dollar investment in a strategically important AI supplier could attract competition-policy attention, particularly because Meta already operates major consumer platforms and AI products.
That risk should be described carefully. This was a minority investment rather than a full acquisition. A potential antitrust concern is not the same as a formal investigation or enforcement action, and the available sources do not establish that regulators blocked or formally challenged the transaction.
Execution risk
Scale’s expertise lies in data production, evaluation, human feedback and AI operations. Those capabilities can support frontier-model development, but they do not automatically produce breakthroughs in model architecture or research.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
Meta therefore had to turn the investment into measurable results: better models, stronger products, improved safety, faster research and effective recruiting. A major stake can accelerate an organization, but it can also create a high financial and managerial hurdle.
Talent-market inflation
The transaction illustrated how expensive AI talent, founder access and strategic positions had become. Paying heavily for a minority stake may be rational if it materially accelerates Meta’s AI program. It also means the investment needs to generate substantial strategic value to justify its cost.
How to judge whether the investment worked
The deal should not be judged solely by whether Meta released another AI model. A more useful evaluation would examine:
- Model performance: Did Meta’s systems improve materially relative to competing models?
- Product adoption: Did Meta AI gain meaningful use across WhatsApp, Instagram, Facebook, Messenger, Threads or Meta’s AI glasses?
- Research output: Did Meta Superintelligence Labs produce credible models, benchmarks or technical advances?
- Data quality: Did Scale’s involvement improve training, evaluation, safety or domain-specific performance?
- Talent retention: Did Meta retain and attract high-value researchers and engineers?
- Customer neutrality: Did Scale preserve relationships with major competing AI companies?
- Financial return: Did the investment create a measurable advantage relative to its roughly $14 billion cost?
Meta’s 2026 announcements provide evidence that its superintelligence organization produced models and agentic features. They do not, by themselves, establish that the Scale transaction caused those outcomes. Multiple factors—including Meta’s computing investment, existing research organization, acquisitions, recruiting and product distribution—could have contributed.
What the deal means for investors and business readers
For investors, the transaction showed that Meta was willing to spend at an extraordinary scale to improve its position in AI. It also showed that the value chain extends beyond model companies and chipmakers.
Data preparation, human feedback, evaluation, safety testing and deployment infrastructure can become strategically important assets. Companies that control or provide those services may influence how quickly AI systems improve, even if they do not build the underlying models.
But the transaction also highlights concentration risk. When a major platform company takes a large stake in a critical supplier, competitors may reassess that supplier’s neutrality. The investment can create value for the investor while making the supplier less attractive to the rest of the market.
For businesses considering Scale or comparable providers, the practical questions are different from the headline valuation. Buyers should assess data security, confidentiality, geographic coverage, human-review quality, evaluation expertise, regulatory requirements, pricing transparency, API integrations and whether the vendor can support a one-time labeling project or a long-term frontier-model program.
Quick wins for a faster PC:
Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →Repair Windows errors before they cause bigger problemsFix Now →Scan for outdated or missing drivers - takes under a minuteDriver Scan →Scale’s enterprise and government orientation may suit complex, high-stakes programs. Smaller teams seeking predictable self-serve pricing or simple image annotation may prefer a software-led or self-hosted alternative. Meta’s investment does not make Scale universally best, and Meta AI is not a substitute for an enterprise data-labeling and evaluation workflow.
The unresolved questions
- Did Meta receive preferential access to Scale’s capabilities without receiving other customers’ confidential data?
- Did Scale retain its major competitors as customers?
- Was the main value in Scale’s data expertise, Wang’s talent, the commercial relationship, strategic influence—or all four?
- Can Meta convert organizational ambition into superior models and widely used products?
- Will regulators view the investment as a meaningful competition concern?
- Will the long-term value of the stake justify the size of the commitment?
Those questions matter more than whether the headline number was $14.8 billion or $14.3 billion. The amount establishes the scale of Meta’s conviction; execution and customer trust determine whether that conviction was justified.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
