AI Wiki

Explore AI Wiki

Citation and evidence

Alexandr Wang

18 min full readUpdated 44 references

This article's verification

Report a problem with this article

More

Use this article

Raw MarkdownExplore connections

Improve this page

Suggest editRevision historyDiscussion

Browse categories

AI CompaniesAI InfrastructurePeople

Cite this article

Alexandr Wang (born January 1997) is an American entrepreneur who co-founded Scale AI in 2016 and, in June 2025, became the first Chief AI Officer of Meta, where he leads the company's Meta Superintelligence Labs. [1][2] He dropped out of the Massachusetts Institute of Technology at age 19 to start Scale AI and built it into one of the largest suppliers of training data and model evaluation for artificial intelligence, work that made him, in 2021 at age 24, one of the youngest self-made billionaires in the world. [3][4][5] In June 2025 Meta invested 14.3 billion dollars for a 49 percent non-voting stake in Scale AI, valuing the company at more than 29 billion dollars, and Wang left the chief executive role to lead Meta's superintelligence effort. [1][6][7] Under Wang, the lab released the proprietary Muse Spark model in April 2026, the terminal coding agent Muse Code in August 2026, and the personal AI agent Muse on September 8, 2026. [22][25][30]

Wang is known for arguing that high-quality human-labeled data is a limiting input for modern AI systems, and for advocating that the United States government invest heavily in the technology to compete with China. [8][9] He has testified before Congress on government adoption of AI and has spoken at international forums on the geopolitics of the field. [10][11] Mark Zuckerberg, Meta's chief executive, has called him "the most impressive founder of his generation." [12]

Who is Alexandr Wang?

Alexandr Wang was born in January 1997 in Los Alamos, New Mexico. [10] His parents were physicists at Los Alamos National Laboratory, the federal weapons research facility, and had immigrated from China. [10][3] He grew up in the small laboratory town, learned the violin, and showed early aptitude in mathematics and computing. [10] As a teenager he competed in national contests, qualifying for the United States Physics Team and reaching the finals of the USA Computing Olympiad. [10]

Wang taught himself to program and worked as a software engineer while still young, including a stint as a programmer at the question-and-answer site Quora and time at the wealth management firm Addepar. [10][13] He was admitted to MIT, where he intended to study mathematics and computer science, and he also held an internship at the high-frequency trading firm Hudson River Trading. [10][13] He attended MIT for about a year before leaving in 2016 to start Scale AI. [10][1]

How did Wang found and build Scale AI?

Wang co-founded Scale AI in 2016 with Lucy Guo, whom he had met while working at Quora. [10][14] The pair, the youngest team in their batch, went through the Y Combinator accelerator and started with 120,000 dollars in seed funding; they initially planned a doctors' concierge service before pivoting to data labeling. [4][14] Guo left in 2018 after differences over the company's direction, while Wang stayed on as chief executive. [14]

Scale's core business was data labeling, the work of annotating images, text, sensor readings, and other inputs so that machine learning models can be trained on them. [10][2] The company combined software tooling with large pools of human annotators and sold the resulting datasets to firms building AI systems. Early customers included companies developing self-driving cars, which needed labeled camera and lidar data. As large language models grew, Scale expanded into producing data for reinforcement learning from human feedback and into evaluating model outputs, work that placed it close to the training pipelines of leading AI labs. [2][15]

Scale also pursued government and defense contracts. The Pentagon's Chief Digital and Artificial Intelligence Office hired the company to help test and evaluate large language models for military use, and Scale won other work with the United States armed forces. [10][16] Wang positioned the company as a contributor to American national security, a theme he returned to in public statements. [8][9]

The company's valuation climbed through several funding rounds. It reached about 7.3 billion dollars in 2021, a level that, with Wang's roughly 15 percent ownership, briefly put his net worth above 1 billion dollars and made him one of the youngest self-made billionaires on record. [10][3][5] In 2024 Scale raised about 1 billion dollars at a valuation near 14 billion dollars, and Wang said he expected revenue to grow sharply. [17] Reporting in 2025 placed the company's annual revenue around 870 million dollars for 2024, with internal projections approaching 2 billion dollars for 2025. [15][17] Forbes estimated Wang's net worth at about 3.6 billion dollars in April 2025. [10] By 2025 Scale counted major AI developers and the United States military among its customers. [15][16]

A prominent young AI executive

As Scale grew, Wang became a visible figure in the AI industry and a commentator on its direction. He was named to the Forbes 30 Under 30 list in enterprise technology, appeared on Time's lists of influential and emerging figures in the field, and spoke at conferences and policy events. [10] During the COVID-19 pandemic he shared a house with Sam Altman, the chief executive of OpenAI, a detail often cited to illustrate his place in the close-knit circle of San Francisco AI leaders. [10]

Wang argued repeatedly that the supply of high-quality data, not only computing power or algorithms, sets the pace of progress in AI, a view that aligned with Scale's business. [2][15] He framed data work as infrastructure for the wider industry rather than as a peripheral service. [2]

What was the June 2025 Meta deal?

In June 2025 Meta reached an agreement to invest 14.3 billion dollars in Scale AI in exchange for a 49 percent stake, a transaction that valued the company at more than 29 billion dollars. [6][7][18] The deal was announced on June 12, 2025. [6] Scale said Meta would hold a minority of its outstanding equity, a Scale spokesperson told CNBC that Meta would not have any voting power, and Reuters reported that Meta did not plan to take a board seat. [1][6][7] As part of the deal Wang stepped down as chief executive of Scale and joined Meta to lead its AI work, while remaining a director on Scale's board. [1][6] Announcing the partnership, Wang said Meta's investment "recognizes Scale's accomplishments to date" and reaffirmed that the company's path forward was "limitless." [1] Scale named Jason Droege, who had joined as chief strategy officer in September 2024 after building and leading companies including Uber Eats, as interim chief executive. [1][6] When Scale named the then Google Cloud chief operating officer Francis deSouza as chief executive on July 30, 2026, effective August 10, it identified Wang as the company's founder and chairman of the board. [35] Wang has also been an independent director of Expedia Group since 2023; stockholders re-elected him at the company's June 17, 2026 annual meeting, with about 79.2 million votes for and 69.8 million withheld. [36][37]

At Meta, Wang became the company's first Chief AI Officer and took charge of a newly created organization named Meta Superintelligence Labs, the unit responsible for Meta AI and the firm's frontier model research. [1][2] According to Fortune, Mark Zuckerberg, Meta's chief executive, had begun meeting regularly with Wang in spring 2025 and described him as "the most impressive founder of his generation." [12] Reporting indicated that Zuckerberg recruited Wang after dissatisfaction with the reception of Meta's Llama 4 model and that he planned to spend heavily on computing infrastructure for the effort. [12] The new lab drew other senior hires, including the former GitHub chief executive Nat Friedman and Daniel Gross of Safe Superintelligence, along with researchers recruited from rival firms, sometimes with large compensation packages. [12]

The deal drew immediate scrutiny over conflicts of interest. Within days, Reuters reported, citing five people familiar with the matter, that Google, Scale's largest customer, planned to cut ties with the company, and that companies competing with Meta were concerned that working with Scale could expose their research priorities to a rival; Google declined to comment. [20] OpenAI also said it was winding down its work with Scale, but said it had begun doing so before the Meta deal was announced. [19] Scale's executives said the company would not share confidential customer information with Meta and that Wang would not be involved in day-to-day operations, but major clients moved away regardless. [19][20] In July 2025 Scale laid off about 200 employees, roughly 14 percent of its staff. [44]

What has Wang led at Meta since 2025?

Wang's first year at Meta combined reorganizations with the lab's first releases. On July 25, 2025, Zuckerberg and Wang announced that Shengjia Zhao, a former OpenAI researcher who had co-founded the new lab, would be its chief scientist, setting its research agenda under Wang's leadership. [39] In October 2025 Meta cut roughly 600 roles in its AI organization, announced in a memo from Wang, affecting its AI infrastructure units, Facebook AI Research (FAIR), and product-related positions. People familiar with the matter told CNBC that TBD Lab, the group of top recent hires that Wang oversees, was spared, and that Meta Superintelligence Labs' workforce was left at just under 3,000. [38]

In March 2026 Meta formed an applied AI engineering organization, led by the Reality Labs vice president Maher Saba and reporting to chief technology officer Andrew Bosworth, to work alongside Meta Superintelligence Labs on the "data engine" used to improve Meta's models, according to an internal memo reported by the Wall Street Journal. [40] When reports described the move as sidelining Wang, Meta spokesperson Andy Stone called them "totally false" and wrote on X that "Alex helped create the new team, still runs MSL and TBD, has growing not waning influence." [41]

The lab's first model, Muse Spark, was released on April 8, 2026. Wang called it "the first model from MSL," wrote that "nine months ago we rebuilt our ai stack from scratch," and said the model now powered Meta AI. [22] Unlike Meta's earlier Llama releases, it was proprietary, although Meta said it hoped to open-source future versions. [23] Muse Spark 1.1 followed on July 9, 2026, together with a public preview of the Meta Model API. [24] On August 5, 2026, Meta released Muse Code in beta, powered by the coding-focused Muse Spark 1.2. [25] Wang told CNBC that Meta was differentiating the tool and its models on price rather than capability against Anthropic and OpenAI, and that a "contributor tier," for developers who opt in to help improve the model, was "more than 10 times cheaper" than pay-as-you-go access. [26] On August 31 he announced that Muse Code was out of beta, with an SDK in developer preview and monthly subscription plans. [27] Meta released Muse Spark 1.3 on September 2, 2026, and Wang announced the public release of its max reasoning setting two days later. [28][29]

Meta launched Muse on September 8, 2026. The company described it as a personal AI agent that runs on a dedicated virtual machine in the cloud, can open a browser and fill out forms, keeps working after the app is closed, and checks with the user before sensitive actions such as sending an email or making a purchase. [30] Wang opened his launch thread on X with "today we're rolling out Muse, our new personal ai assistant," and described it as "always-on, wicked fast" and "designed to be secure." [31] Over the following two weeks he used his X account to announce an expanded beta of outbound phone calls (September 16), a partnership with Shopify "to enable agentic checkout with Shop Pay on all Shopify stores" (September 21), and a partnership with PayPal covering "all PayPal merchants worldwide" (September 22). [32][33][34] Expedia, on whose board Wang sits, announced on September 22 that it was also joining Muse. [43]

DateEvent at Meta under Wang
June 2025Joins Meta as its first Chief AI Officer, leading Meta Superintelligence Labs [1][2]
July 25, 2025Shengjia Zhao named MSL chief scientist [39]
October 22, 2025About 600 AI roles cut, announced in a memo from Wang [38]
March 2026Applied AI engineering organization formed under CTO Andrew Bosworth; Meta denies reports that Wang was sidelined [40][41]
April 8, 2026Muse Spark, MSL's first model, released [22][23]
July 9, 2026Muse Spark 1.1 and public preview of Meta Model API [24]
August 5, 2026Muse Code (beta) and Muse Spark 1.2 [25][26]
August 31, 2026Muse Code leaves beta [27]
September 2 and 4, 2026Muse Spark 1.3, then its max reasoning setting [28][29]
September 8, 2026Muse personal agent launched [30][31]
September 21 and 22, 2026Shopify and PayPal partnerships announced for Muse [33][34]

Expanded article table

What are Wang's views on AI and US-China competition?

Wang has been an outspoken advocate for American leadership in artificial intelligence. In January 2025 he placed a full-page open letter addressed to President Donald Trump in a national newspaper, declaring that America must win the AI war and setting out a multi-point plan for the federal government to invest more in the technology. [8][9] He attended Trump's second inauguration that month. [10]

At the World Economic Forum in January 2025 Wang spoke about the competition between the United States and China, focusing on the Chinese developer DeepSeek and its release of the DeepSeek-R1 reasoning model. [11][21] He said the model showed that China had rapidly narrowed the gap with American systems and estimated, in widely quoted remarks, that DeepSeek had access to a large fleet of advanced chips, comments he used to question the effectiveness of United States export controls. [21] He disputed the idea that China held an unbeatable advantage in data, arguing that the United States government, especially in national security, held large datasets that could become a source of advantage if used. [11] He also called for the United States to expand energy production to power AI computing and urged the government to adopt AI more aggressively in its own operations. [11]

Reception

Colleagues and investors have described Wang as an unusually capable recruiter and commercial operator. The investor Sarah Guo called him exceptionally smart and ambitious, and former Scale managers credited his skill at building teams and winning customers. [12] Observers also noted that he is an entrepreneur rather than a research scientist, and some questioned whether a non-researcher could lead and inspire the academic talent inside Meta's AI organization. [12] The most prominent such criticism came from Yann LeCun, who announced in November 2025 that he was leaving Meta. In a Financial Times interview reported by CNBC on January 5, 2026, LeCun called Wang "young" and "inexperienced"; he said Wang "learns fast" and "knows what he doesn't know," but that "there's no experience with research or how you practice research." [42]

The scale of the Meta investment and Wang's compensation prompted debate about how much value individual AI leaders command and about the consolidation of talent and capital among a few large companies. [12][6] The departure of Scale's customers after the deal was cited as an example of how closely the major AI developers guard their data and research, and of the difficulty of remaining a neutral supplier once a rival holds a large equity stake. [19][20]

Facts

ItemDetail
Full nameAlexandr Wang
BornJanuary 1997, Los Alamos, New Mexico, United States
ParentsPhysicists at Los Alamos National Laboratory
EducationLos Alamos High School; attended MIT (mathematics and computer science), left at age 19 after about one year
Known forCo-founding Scale AI; leading Meta Superintelligence Labs
Company co-foundedScale AI (2016, with Lucy Guo, via Y Combinator)
Role at Scale AIChief executive 2016 to 2025; director after June 2025; described by Scale as chairman of the board in July 2026
Role at MetaChief AI Officer; head of Meta Superintelligence Labs (from June 2025)
Major releases at MetaMuse Spark (April 2026), Muse Code (August 2026), Muse (September 2026)
Other board rolesExpedia Group, independent director since 2023
Meta investment in Scale AIAbout 14.3 billion dollars for a 49 percent non-voting stake; Scale valued at more than 29 billion dollars
Net worth milestoneSurpassed 1 billion dollars in 2021 at age 24; Forbes estimated about 3.6 billion dollars in April 2025
RecognitionForbes 30 Under 30; Time lists of influential figures in AI

Expanded article table

References

  1. ^1 ^2 ^3 ^4 ^5 ^6 ^7 ^8 ^9"Founder, Alexandr Wang, Joins Meta to Work on AI Efforts." Scale AI. scale.com/...unces-next-phase-of-company-evolution
  2. ^1 ^2 ^3 ^4 ^5 ^6 ^7"Alexandr Wang, Chief AI Officer." Meta. meta.com/...alexandr-wang
  3. ^1 ^2 ^3"Alexandr Wang: how the world's youngest self-made billionaire is shaping the future of AI." The Week. theweek.com/...alexandr-wang-profile
  4. ^1 ^2"Founder Story: Alexandr Wang of Scale AI." Frederick AI. frederick.ai/...alexandr-wang-scale-ai
  5. ^1 ^2"Who Is Alexandr Wang, the Founder of Scale AI Joining Meta?" Entrepreneur. entrepreneur.com/...493281
  6. ^1 ^2 ^3 ^4 ^5 ^6 ^7"Scale AI's Alexandr Wang confirms departure for Meta as part of 14.3 billion dollar deal." CNBC. cnbc.com/...-exit-for-meta-part-of-14-billion-deal
  7. ^1 ^2 ^3"Meta poaches 28-year-old Scale AI CEO after taking multibillion dollar stake in startup." Reuters (Yahoo Finance syndication), June 13, 2025. finance.yahoo.com/...investment-scale-ai-013403682
  8. ^1 ^2 ^3"Scale AI CEO Alexandr Wang to Trump: America must win the AI war." Semafor. semafor.com/...o-trump-america-must-win-the-ai-war
  9. ^1 ^2 ^3"America Must Win the AI War." Scale AI. scale.com/...win-the-ai-war
  10. ^1 ^2 ^3 ^4 ^5 ^6 ^7 ^8 ^9 ^10 ^11 ^12 ^13 ^14 ^15 ^16"Alexandr Wang." Wikipedia. en.wikipedia.org/...Alexandr_Wang
  11. ^1 ^2 ^3 ^4"Alexandr Wang on the AI Race with China." The AI Insider. theaiinsider.tech/...ang-on-the-ai-race-with-china
  12. ^1 ^2 ^3 ^4 ^5 ^6 ^7"Alexandr Wang is now leading Meta's AI dream team. Will Mark Zuckerberg's big bet pay off?" Fortune. fortune.com/...ale-ai-entrepreneur-mark-zuckerberg
  13. ^1 ^2"Who Is Alexandr Wang, the Founder of Scale AI Joining Meta?" Entrepreneur. entrepreneur.com/...493281
  14. ^1 ^2 ^3"Lucy Guo." Wikipedia. en.wikipedia.org/...Lucy_Guo
  15. ^1 ^2 ^3 ^4"Scale AI: The infrastructure behind OpenAI, Meta, and the U.S. Military." Today in AI. todayin-ai.com/...scale
  16. ^1 ^2"Scale AI to set the Pentagon's path for testing and evaluating large language models." DefenseScoop. defensescoop.com/...aluating-large-language-models
  17. ^1 ^2"Scale AI secures 1 billion dollar funding at 14 billion dollar valuation." Fortune. fortune.com/...ion-ceo-alexandr-wang-profitability
  18. ^"Meta ramps up AI efforts by paying 14.3 billion dollars for 49 percent of Scale." Fortune. fortune.com/...lexandr-wang-superintelligence-team
  19. ^1 ^2 ^3"OpenAI drops Scale AI as a data provider following Meta deal." TechCrunch. techcrunch.com/...ata-provider-following-meta-deal
  20. ^1 ^2 ^3"Google, Scale AI's largest customer, plans split after Meta deal." CNBC. cnbc.com/...t-customer-plans-split-after-meta-deal
  21. ^1 ^2"Scale AI CEO says China has quickly caught the U.S. with the DeepSeek open-source model." CNBC. cnbc.com/...as-quickly-caught-the-us-with-deepseek
  22. ^1 ^2 ^3Alexandr Wang (@alexandr_wang). "1/ today we're releasing muse spark, the first model from MSL." X post, April 8, 2026. x.com/...2041909376508985381
  23. ^1 ^2Jonathan Vanian. "Meta debuts new AI model, attempting to catch Google, OpenAI after spending billions." CNBC, April 8, 2026. cnbc.com/...billion-deal-to-bring-in-alexandr-wang
  24. ^1 ^2Meta Superintelligence Labs. "Introducing Muse Spark 1.1." Meta AI Research, July 9, 2026. research.meta.ai/...cing-muse-spark-meta-model-api
  25. ^1 ^2 ^3Meta Superintelligence Labs. "Introducing Muse Code and Muse Spark 1.2." Meta AI Research, August 5, 2026. research.meta.ai/...g-muse-code-and-muse-spark-1-2
  26. ^1 ^2Jonathan Vanian. "Meta debuts first AI coding agent to take on Anthropic and OpenAI." CNBC, August 5, 2026. cnbc.com/...-code-to-take-on-anthropic-and-openai-
  27. ^1 ^2Alexandr Wang (@alexandr_wang). "1/ muse code is out of beta." X post, August 31, 2026. x.com/...2094502557129543774
  28. ^1 ^2Meta Superintelligence Labs. "Introducing Muse Spark 1.3." Meta AI Research, September 2, 2026. research.meta.ai/...introducing-muse-spark-1-3
  29. ^1 ^2Alexandr Wang (@alexandr_wang). "1/ we just publicly released Muse Spark 1.3 max!" X post, September 4, 2026. x.com/...2095938990197329935
  30. ^1 ^2 ^3Meta. "Introducing Muse: The World's First Personal AI Agent Built for Everyone." Meta Newsroom, September 8, 2026. about.fb.com/...introducing-muse-personal-ai-agent
  31. ^1 ^2Alexandr Wang (@alexandr_wang). "1/ today we're rolling out Muse, our new personal ai assistant." X post, September 8, 2026. x.com/...2097402344061510004
  32. ^Alexandr Wang (@alexandr_wang). Post announcing an expanded Muse phone-call beta, quoting Ryan Fox on outbound calls to US businesses. X post, September 16, 2026. x.com/...2100349174529265967
  33. ^1 ^2Alexandr Wang (@alexandr_wang). "we are excited for muse to be partnering deeply with @Shopify to enable agentic checkout with Shop Pay on all Shopify stores!" X post, September 21, 2026. x.com/...2102092011021217911
  34. ^1 ^2Alexandr Wang (@alexandr_wang). "we're partnering with @PayPal to enable customers to seamlessly shop and checkout online with @Muse across all PayPal merchants worldwide." X post, September 22, 2026. x.com/...2102418455622508722
  35. ^Scale AI. "Scale AI Appoints Francis deSouza as CEO to Lead Next Phase of Company's Growth." July 30, 2026. scale.com/...scale-appoints-new-ceo
  36. ^Expedia Group. "2026 Proxy Statement." April 2026. s202.q4cdn.com/...EXPE2026ProxyStatement_Web.pdf
  37. ^Expedia Group. Form 8-K, Item 5.07 (results of the 2026 Annual Meeting of Stockholders). U.S. Securities and Exchange Commission, June 2026. sec.gov/...expe-20260617
  38. ^1 ^2Ashley Capoot. "Meta lays off 600 from 'bloated' AI unit as Wang cements leadership." CNBC, October 22, 2025. cnbc.com/...meta-layoffs-ai
  39. ^1 ^2Maxwell Zeff. "Meta names Shengjia Zhao as chief scientist of AI superintelligence unit." TechCrunch, July 25, 2025. techcrunch.com/...ist-of-ai-superintelligence-unit
  40. ^1 ^2"Meta creates new applied AI engineering division." The Decoder, March 4, 2026. the-decoder.com/...applied-ai-engineering-division
  41. ^1 ^2"Meta Shuts Down Rumors Mark Zuckerberg Sidelined Chief AI Officer Alexandr Wang, Spokesperson Calls Report False: 'This Is So Silly'." Benzinga (Yahoo Finance syndication), March 11, 2026. finance.yahoo.com/...ts-down-rumors-mark-133048348
  42. ^Sawdah Bhaimiya. "AI godfather says Meta's new 29-year-old AI boss is 'inexperienced' and warns of staff exodus." CNBC, January 5, 2026. cnbc.com/...-ai-boss-alexander-wang-inexperienced-
  43. ^Expedia (@Expedia). "Here at Expedia, we love planning trips. Soon, your personal AI agent will too." X post, September 22, 2026. x.com/...2102481062429426085
  44. ^"Scale AI lays off 14% of staff, largely in data-labeling business." TechCrunch, July 16, 2025. techcrunch.com/...argely-in-data-labeling-business

Improve this article

Add missing citations, update stale details, or suggest a clearer explanation. Every suggestion is reviewed for sourcing before it goes live.

5 revisions · v6 · 3,558 words · full history

Fact-checks are independent of edits: a reviewer re-verifies the article against its sources and stamps the date. How we verify

Research and drafting on this wiki are AI-assisted, under named human editorial standards. How AI is used here

Reviewer note: xg06 V13 independent verification of 2025-2026 Meta tenure, Scale and Expedia roles (quotes checked word-for-word); 1 material + 3 minor fixed 2026-09-23

Cite this page: AI Wiki. "Alexandr Wang." aiwiki.ai, updated 28 Sept 2026, fact-checked 23 Sept 2026. CC BY 4.0. https://aiwiki.ai/wiki/alexandr_wang

Suggest edit