Do these 3 things before closing this tab:
1Fix the driver behind crashes, sound loss and screen glitches2Repair Windows errors before they cause bigger problems3Scan for outdated or missing drivers - takes under a minuteSome links on this page are affiliate links: if you buy through them we may earn a commission, at no extra cost to you.
Meta’s investment in Scale AI did trigger a serious trust shock. The roughly $14.3 billion deal gave Meta an approximately 49% minority stake in Scale, moved Scale founder Alexandr Wang to Meta, and prompted reported customer defections or diversification. But “customer exodus” is too absolute, and there is no public evidence proving that Mercor captured a quantified windfall.
The more durable story is a change in how AI companies assess data vendors: neutrality, information barriers, expert quality, and control over sensitive evaluation data now matter alongside price and scale.
What Meta actually bought
Meta did not simply acquire Scale AI outright. Scale announced that Meta invested approximately $14.3 billion for a minority stake of about 49%, valuing Scale at more than $29 billion. Scale said it would remain an independent company and continue serving other customers. Alexandr Wang, Scale’s founder and chief executive, moved to Meta to help lead its artificial-intelligence efforts.
Scale’s announcement describes the transaction as a new phase for the company, not a 100% acquisition. That distinction matters. A minority, reportedly non-voting stake does not automatically give Meta operational control, access to competitors’ information, or the right to direct every Scale decision.
#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.
It also matters financially. The headline $14.3 billion figure represents the value of the investment transaction; it should not automatically be described as money deposited into Scale’s operating budget. A transaction can include purchases from existing shareholders, new capital for the company, or both. The public reporting supplied for this story does not provide a complete breakdown of cash paid to shareholders, primary capital invested in Scale, or Wang’s separate compensation arrangements.
What Meta obtained was a large economic interest in an important AI infrastructure supplier, plus Scale’s founder in a senior role at Meta. Even without voting control, that combination was enough to raise commercial and governance questions.
Why customers questioned Scale’s neutrality
Scale provides data-labeling, training-data, and model-evaluation services. Those workflows can involve proprietary prompts, evaluation rubrics, model outputs, failure cases, security information, and other details that reveal how an AI company is developing and testing its systems.
Free tools Windows power users keep installed
One-click scans. No signup required.
Many of Scale’s customers compete with Meta, including major technology companies and frontier AI labs. After the investment, those customers had to consider whether they wanted a strategic rival to hold a large economic stake in their data and evaluation supplier.
The concern did not require evidence of misconduct. It was a vendor-risk issue involving:
- Confidentiality: Could sensitive customer information be kept inaccessible to Meta?
- Data segregation: Were systems, staff, permissions, and workflows separated effectively?
- Competitive intelligence: Could evaluation methods or model weaknesses reveal strategic information?
- Preferential access: Would Meta receive better service, faster capacity, or indirect insight?
- Dependence: Would customers become too reliant on a company backed by a direct competitor?
- Contract rights: Could a customer terminate or renegotiate after a major ownership and leadership change?
Scale said it remained independent and was committed to safeguarding customer data. That is an important company assurance, but it is not independent proof that every customer’s concerns were resolved.
Which customers actually moved?
Google: a planned split, followed by some resumed work
Reuters reported that Google, described as Scale’s largest customer, planned to cut or sharply reduce its relationship with Scale after the Meta transaction. The report also said many AI laboratories were considering building more labeling and data-operation capabilities internally.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
However, later Forbes reporting said Google initially moved away from Scale but resumed some work a few months later. The careful description is therefore that Google planned a split or reduced dependence—not that Google permanently abandoned Scale.
Source: Reuters, republished by Investing.com.
OpenAI: reported departure
Forbes later reported that OpenAI dropped Scale after Meta’s investment. This remains secondary reporting attributed to people familiar with the matter rather than a public confirmation from OpenAI or Scale. It is reasonable to describe OpenAI as a reported customer loss, but not as independently verified public fact.
Microsoft and other customers
Scale has served other major technology companies, but the available public evidence does not establish the post-deal status of every customer. It would be inaccurate to imply that Microsoft, Amazon, every frontier lab, or the entire customer base left.
The evidence supports a narrower conclusion: some important customers reassessed Scale, moved some work, diversified suppliers, or considered internal alternatives.
Quick wins for a faster PC:
Clear out junk files and repair common Windows errorsFree Scan →Scan for outdated or missing drivers - takes under a minuteDriver Scan →Did Scale lose business?
Possibly—but customer loss and total revenue are not the same thing.
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.
Forbes reported that Scale generated just under $1 billion in revenue in 2025, compared with approximately $870 million the previous year. Forbes also reported that Meta agreed to pay Scale at least $450 million annually for five years, or more than half of Meta’s annual AI spending, whichever was less.
Those figures are attributed to Forbes and are not presented here as audited public-company results. If accurate, the arrangement could cushion or outweigh revenue lost from rival customers in the short term.
This creates three separate questions:
| Question | What the evidence suggests |
|---|---|
| Did Scale lose some customers or projects? | Reportedly yes, including a reported OpenAI departure and Google’s planned reduction. |
| Did Scale’s total revenue collapse? | No. Forbes reported revenue growth in 2025. |
| Did Scale’s strategic position weaken? | Potentially. Losing neutral third-party business can increase dependence on Meta even if revenue rises. |
A company can become financially stronger while becoming less acceptable to customers that need a neutral supplier. Conversely, a rival can gain strategic importance without immediately disclosing large revenue gains.
Why Mercor became a plausible beneficiary
Mercor’s opportunity is broader than replacing commodity image labeling. Its public offering includes expert recruitment, human-data collection, model training, model evaluation, enterprise-agent evaluations, and license-ready expert-produced datasets.
Mercor says its enterprise customers can source and vet specialists, embed experts into workflows, use managed evaluations, or operate through a self-serve evaluation platform. Its public materials also emphasize customer control over tooling, visibility into annotator quality and workflows, and an “open-box” alternative to less transparent operations.
Those capabilities may be especially relevant as AI development moves toward difficult, judgment-heavy work in fields such as coding, science, law, finance, reasoning, and agent behavior.
Mercor’s potential advantages
- Perceived neutrality: Mercor does not carry the same reported Meta ownership connection as Scale.
- Specialist access: An expert network can be useful for evaluating advanced models where generalist labeling is insufficient.
- Flexible engagement: Mercor promotes hourly and cost-plus structures as well as managed programs.
- Data-control options: Mercor says customers can use their own platform or Mercor’s tooling.
- Speed: Mercor says some enterprise workflows can reach production in four to six weeks.
Mercor’s official information is available through its enterprise partnership page, enterprise evaluations page, and dataset offering.
PC 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 & 11Crashes, 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 minuteIs Mercor’s “windfall” real?
There are different levels of evidence.
Strongest evidence: TechCrunch reported that researchers at Meta’s new AI organization preferred Surge and Mercor over Scale for some work. Mercor has also expanded its public positioning around expert networks, evaluations, and frontier-AI data.
Plausible inference: Some Scale customers likely tested or shifted work to Mercor because of neutrality, confidentiality, or data-control concerns. Mercor’s expert-heavy model may also have benefited from the industry’s shift toward more sophisticated evaluations.
Not publicly established: There is no disclosed figure showing how much revenue Mercor gained from former Scale customers. The public evidence does not identify every customer that switched, quantify the size of those contracts, or prove that Mercor’s growth came primarily from Scale’s disruption rather than from overall AI-market expansion.
Mercor should therefore be described as a strategic beneficiary or likely beneficiary—not as a proven recipient of hundreds of millions of dollars in displaced Scale revenue.
Mercor is not the only alternative
Labelbox
Reuters reported that Labelbox’s chief executive expected the company to generate hundreds of millions of dollars in new revenue by year-end from customers leaving Scale. That is an executive forecast, not audited realized revenue. Labelbox remains an alternative to evaluate for labeling and annotation workflows; its official site is Labelbox.com.
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.
Surge AI
TechCrunch’s reporting identified Surge AI alongside Mercor as a company preferred for some work. Surge’s official site is SurgeHQ.ai. Current public pricing and the precise scale of any customer transfers are not established by the supplied evidence.
Internal data teams
Some AI companies may bring labeling, evaluation, recruiting, and quality assurance in-house. This offers maximum control over sensitive data and proprietary evaluation methods, but it requires recruiting, tooling, compliance, workforce management, and enough volume to justify the fixed cost.
The likely outcome is not one universal winner. The market may fragment among large-scale vendors, expert networks, specialist providers, and internal teams.
The Scale–Mercor lawsuit adds risk to the hiring race
Scale sued Mercor in September 2025, alleging trade-secret theft. The existence of the lawsuit is verifiable, but the allegations are disputed and have not been established as adjudicated facts. Axios reported on the case.
The dispute highlights a broader operational risk. When a customer shock creates a hiring rush, recruiting former employees can transfer valuable experience quickly—but it can also create allegations involving confidential information, restrictive covenants, customer lists, or internal processes.
Any company considering Mercor should examine employee-transition controls, confidentiality procedures, access permissions, source-of-data records, and indemnification terms. A vendor’s ability to hire quickly is useful only if it can do so without creating legal or provenance problems.
What enterprise buyers should evaluate
Customers deciding whether to stay with Scale, add Mercor or another vendor, or build internally should score the options against the same practical criteria:
The Tool Desk
Outbyte Driver Updater FREEFix the driver behind crashes, sound loss and screen glitchesFind Drivers →Outbyte PC Repair FREEClear out junk files and repair common Windows errorsFree Scan →- Ownership and governance: Identify investors, voting rights, board influence, and change-of-control provisions.
- Neutrality: Determine whether the provider serves direct competitors and how conflicts are managed.
- Information barriers: Require specific answers about systems, personnel, permissions, encryption, retention, deletion, and audit rights.
- Data provenance: Document where workers, examples, labels, and evaluation results came from.
- Expert quality: Test domain qualifications, reviewer calibration, disagreement handling, and quality measurement.
- Commercial model: Compare per-task, hourly, cost-plus, platform, and dataset-license costs on a fully loaded basis.
- Turnaround and scale: Confirm how quickly the vendor can staff a project and sustain production volume.
- Exit options: Ensure the customer can export data, terminate work, and transition to another provider.
- Continuity: Assess what happens if ownership, leadership, workforce availability, or customer concentration changes again.
Scale may fit when
Scale remains relevant for large production programs, established infrastructure, and customers that value existing capacity and relationships. It may be a poor fit for an organization whose central requirement is avoiding any strategic supplier relationship with Meta.
Mercor may fit when
Mercor’s public positioning is strongest for expert-driven data, model evaluation, and higher-judgment work. It may be less suitable for enormous volumes of standardized, low-complexity labeling where the lowest unit cost is the overriding objective. Enterprise pricing is sales-led; public contractor rates are not customer invoices.
As examples, public Mercor listings have shown rates such as $50 per hour for generalist AI-training work and $75–$100 per hour for some writing roles. Rates and availability can change, and a worker’s compensation is not the same as a buyer’s fully loaded cost. See Mercor’s expert page for current public opportunities.
The unresolved question: can Scale remain neutral?
Scale can remain legally independent while becoming commercially less neutral in customers’ eyes. The practical issue is not only whether Meta controls Scale. It is whether rival customers are comfortable entrusting strategic data to a provider in which Meta owns a large economic stake and whose founder joined Meta.
Recommended Free Tools
That distinction explains why formal safeguards may not fully restore confidence. A contract can restrict access, but customers may still diversify to reduce dependence, protect negotiating leverage, and avoid future ownership surprises.
For Scale, Meta’s spending commitment may provide financial support and a major customer relationship. For Scale’s rivals, the investment creates an opening. For buyers, it is a reminder that vendor concentration and ownership structure belong in AI procurement reviews—not just cybersecurity questionnaires.
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.
