Nansen
Nansen: Block explorer or on-chain query tool for addresses, transactions, contracts, PoW chains, and multichain activity.
ABAB Structured Brief
Nansen is indexed in ABAB Crypto Map under Block Explorers & On-chain Tools. This page keeps the official site, category, tags, and related ABAB coverage together as a searchable crypto project profile. Official domain: nansen.ai.
Related News & Analysis
Nansen CEO: AI Agents Will Surpass Human Trading Volume by 2028
Nansen CEO Alex Svanevik stated that AI agents will exceed human trading volume by 2028. He expects that within about two years, the number of AI trading agents will surpass human traders. Nansen is shifting from on-chai...
From Norwegian Special Forces to a Global “Day Zero” Venture Machine: Antler, Magnus Grimeland, and the Architecture of Global Startup Investing
First, the classification: Antler is indeed a venture-capital organization, but it is more accurate to describe it as a global early-stage VC firm and multi-fund platform rather than a single “venture fund.” Antler describes itself as a global early-stage VC firm and a “Day Zero investor.” Its distinguishing idea is to engage founders extremely early—sometimes before a company exists, before the founding team is complete, and even before the founder has settled on a specific idea. It manages multiple regional early-stage funds and has expanded into later-stage investing through Antler Elevate. The Founder Residency, co-founder matching and business-model validation resemble an accelerator or company builder, but capital allocation remains the core function. Antler’s current Fact Sheet says more than 70 Partners raise and invest funds; its 2025 Nordic Fund II announcement described a global network of more than 15 funds. The best description is therefore: a global VC platform vertically integrating talent discovery, company formation, very-early-stage equity investing and follow-on capital. Antler was created in 2017, began investing in 2018, and originated in Singapore. Magnus Grimeland began building Antler in 2017, while Antler’s current Fact Sheet identifies 2018 as the year it began investing. Its first Founder Residency was also launched in Singapore in 2018. The precise current city count is inconsistent across Antler’s own public pages. The main About text says 27 cities, Magnus’s biography on the same site says 30, and the Fact Sheet says more than 30. The safest formulation is therefore that Antler operates across roughly 30 cities worldwide. Current official materials say Antler has backed more than 1,800 companies. Magnus’s 2025 retrospective said AUM had grown from roughly $5 million in 2018 to more than $1 billion. AUM, however, is capital managed on behalf of investors; it should not be confused with assets personally owned by Antler or Magnus. The official founder count is five, not one. Antler’s current Fact Sheet identifies five co-founders: Magnus Grimeland, Fridtjof Berge, Dilan Mizrakli-Landgraff, Vegard Medbø, and Jussi Salovaara. Magnus nevertheless dominates the public origin story because he is Founder & CEO, chief strategic figure and principal spokesperson. Media coverage therefore often describes him simply as “the founder of Antler,” even though Antler itself now formally recognizes five co-founders. The division of labor among those founders shows that Antler was not simply a personal investment vehicle for Magnus. Magnus is Founder & CEO and the central public and strategic leader. Fridtjof Berge is Co-founder & Chief Business Officer, overseeing external activities and serving on investment committees. Dilan Mizrakli Landgraff is Co-founder & Chief Strategy Officer with experience in investment decisions, acquisitions, exits and financing. Vegard Medbø is Co-founder & COO and has played a major role in operations and geographic expansion. Jussi Salovaara is Co-founder & Managing Partner Asia and leads APAC investments and founder support. Magnus is best understood as the system architect and brand center, while the other founders provide investment, operating, institutional and regional infrastructure. Antler’s creation is partly a story of a McKinsey alumni network turning into an entrepreneurial-capital network. McKinsey’s own alumni material lists Magnus in Oslo from 2007 to 2013 and Fridtjof Berge from 2012 to 2017. Berge said the two reconnected in 2017 after both had left McKinsey, began discussing what they should do next, and quickly aligned around creating a company capable of producing large-scale positive impact. This helps explain one of Antler’s defining characteristics: its original resource base was not merely money. It was a dense network accumulated through Harvard, McKinsey, Rocket Internet/Zalora, Asian technology ecosystems and Nordic business circles, later extended to institutional asset managers, sovereign capital, pensions and global VC firms. Magnus’s early environment was rural Norway rather than Silicon Valley. Singapore Global Network reports that he grew up on a husky farm in Sande, Norway. Magnus himself recalls exploring forests as a child and being fascinated by Norwegian explorers such as Fridtjof Nansen, Roald Amundsen and Thor Heyerdahl. That background is consistent with the language he later used around founders: adventure, endurance, ambition and the belief that exceptional ability can emerge from non-traditional environments. The most revealing publicly documented aspect of his family environment is the contrast in values represented by his parents. Magnus has said that the death of his great-grandmother when he was 11 triggered an intense interest in the meaning of life. He described his father as more socialist in outlook; his father directed him toward Kafka’s The Trial and other major books. His mother, in his account, offered a more spiritual answer. He also said his family had a multigenerational tradition of national service, helping explain why he chose the military after international boarding school rather than immediately entering university. His first major deliberate move out of his original resource environment was United World College. Magnus recalls discovering an article about Atlantic College in Wales when he was roughly 12. Norway’s Queen Sonja was visiting, and Nelson Mandela—whom Magnus admired—was associated with the institution. He immediately told his teachers he wanted to attend, only to learn that he had to wait until 16 and needed strong grades and accomplishments. He eventually received a United World College scholarship and attended Atlantic College. Singapore Global Network independently confirms the scholarship. His LinkedIn history lists UWC Atlantic College from 1997 to 1999. Magnus later said he initially struggled because it was his first time studying entirely in English and many international classmates were academically ahead of him. The key effect was not merely academic. It placed someone from rural Norway inside a high-density global talent network for the first time. After UWC, he joined Norway’s naval special forces rather than going directly to university. Forbes reports a roughly two-year stint. Magnus identifies the unit as the Norwegian Naval Special Forces, Marinejegerkommandoen, and credits the experience with building mental and physical rigor, purpose and humility. That experience also provides context for Antler’s later organizational language. “Uncomfortably ambitious” is now one of Antler’s formal guiding principles, alongside unusually strong emphasis on resilience and execution. In 2003, at age 23, Magnus entered Harvard at an unusually consequential moment in Internet history. Forbes says he arrived as a 23-year-old freshman in 2003. His LinkedIn profile lists a Bachelor of Arts in Economics, 2003–2007, while Antler’s official profile confirms he is a Harvard graduate. Singapore Global Network notes that his Harvard years coincided with broadband adoption, the emergence of social media and the rise of major Internet platforms. Harvard therefore gave him more than economics training: it placed him inside the social network of a generation that was beginning to build and finance the modern Internet economy. Those Harvard relationships later acquired direct economic value. Forbes reports that Magnus befriended Facebook co-founder Eduardo Saverin and met Cameron and Tyler Winklevoss through rowing. Singapore Global Network also names Joshua Kushner, later founder of Thrive Capital, and Sam Teller, later Elon Musk’s chief of staff, among his broader Harvard-era circle. Saverin later became an investor in Antler. Forbes reported that by 2022 Antler’s backers included Schroders, the International Finance Corporation and Saverin. This reveals a recurring Magnus pattern: education network → professional network → founder network → LP network. He came close to an early Facebook opportunity but never joined, and later turned that near-miss into a broader theory of talent proximity. Forbes reports that Magnus considered pursuing an internship at the early TheFacebook but was balancing academics, athletics and caring for an infant son. His later conclusion was not simply that he regretted missing Facebook. Instead, he argued that the more important strategy is to remain near unusually creative and capable people so that exceptional opportunities recur. Antler later turned that concept into a product: artificially creating talent density and repeatedly placing potential founders around one another. Family responsibilities were present long before he became professionally successful. Forbes confirms that Magnus was caring for an infant son while at Harvard. In a 2025 Singapore Global Network interview, he referred to his eldest son having attended high school in Singapore and to two younger children being raised there. That context also helps explain his first post-Harvard decision. Magnus has said that he graduated with student loans and needed stable income, making McKinsey an easy choice when the opportunity arrived. After Harvard, he also worked on a development project in Zambia, reinforcing the theme of unequal access to opportunity. Magnus recalls traveling to northern Zambia to establish a development project; Antler’s interview material includes a photograph of him teaching first aid there in 2007. He connected the lack of clean water, electricity, healthcare and educational opportunity he saw there with disparities he later encountered in Asia. His broader conclusion became that the world does not lack talented people; it lacks equal access to the networks and capital that allow talent to compound. That idea eventually became one of Antler’s deepest underlying investment theses. McKinsey was his first major representative business career and functioned primarily as a training ground. McKinsey’s alumni records place Magnus in Oslo from 2007 to 2013. Other professional profiles describe roughly six years at the firm, ultimately at Junior Partner/partner level, working around technology, telecom and media across Europe, North America and Asia. Magnus himself has described McKinsey as an almost ideal training ground where he could work closely with major technology businesses while building a stable financial base. Singapore Global Network says the experience also produced his early conviction about Asia: unusually strong talent combined with rapidly expanding markets. The move from McKinsey to Zalora was the decisive transition from analyzing companies to building one. The precise year is inconsistent across public accounts. Singapore Global Network says Magnus moved to Singapore and joined Zalora as a co-founder in 2012; other media commonly place the transition in 2013, while McKinsey lists his tenure through 2013. The safest formulation is that the transition occurred around 2012–2013. Zalora was a Southeast Asian fashion e-commerce company associated with Rocket Internet. Magnus helped develop its marketplace and establish local operations across the region. At the time, payments and logistics infrastructure remained underdeveloped in several markets, forcing the company to build significant operating infrastructure itself. Zalora’s most important lesson for Magnus was not fashion retail; it was how to scale a company in an immature operating environment. Zalora was eventually consolidated into Global Fashion Group, where Magnus became COO. Antler’s current biography says he helped scale GFG across 26 markets. Even more important was the entrepreneurial spillover. People around the Zalora ecosystem subsequently helped build or lead companies such as Gojek, Shopee, StashAway and ShopBack. Both Singapore Global Network and Magnus’s own interviews identify this “startup mafia” dynamic as an important precursor to Antler. He saw that one of the most valuable by-products of a high-growth company could be a cohort of operators who already knew how to build companies quickly. Why leave fashion e-commerce? Magnus’s own answer is unusually direct: fashion was not his deepest passion. He has said that he did not wake up every morning thinking about fashion. What continued to occupy him was the question of how to enable more exceptional people to become founders. At McKinsey, he had seen engineers who seemed underutilized inside large companies; at Zalora, he saw former colleagues leave and successfully build companies of their own. These observations converged into the original Antler questions: Who might become an exceptional founder? How can that person find the right co-founder? How can a team identify a real problem quickly? How can an investor supply the first capital before the company is fully formed? Those questions remain central to Antler’s own account of why it was founded. Choosing Singapore rather than the United States or Europe was another major strategic decision. Magnus has said he compared several possible bases before choosing Singapore because it was easy to establish and operate a business there, it provided access to a rapidly digitizing region, government-business cooperation was unusually strong, and it was attractive for long-term family life. The long-term significance was substantial. Had Antler originated in Silicon Valley, it could easily have become another Silicon Valley seed fund. Beginning in Singapore forced its claim that “innovation is global” to become an operating problem rather than merely a slogan. In a 2024 TechCrunch interview, Magnus continued to describe Asia as one of the largest global opportunities for entrepreneurial growth, highlighting Southeast Asia, India, Japan and Korea. Organization, capital, business model, results, criticism and final assessment Antler did not begin with a giant institutional fund; it began with personal capital plus relationship capital. Forbes reported that Magnus initially committed about $500,000 of his own capital in 2017, generated from his previous fashion-industry career, before raising roughly $6 million from colleagues and fellow entrepreneurs in 2018. Magnus’s 2025 retrospective separately characterized 2018 AUM as approximately $5 million. The accounting scopes are not identical, but both demonstrate that the first Antler operation was tiny relative to its eventual institutional scale. The first 2018 Founder Residency already contained nearly the entire model that Antler later scaled globally. Magnus says the team conducted roughly 700 conversations before bringing 62 people into the first Singapore residency in July 2018. Forbes gives another, compatible metric: roughly 1,400 applicants, 62 accepted. One number measures conversations, the other applications. After roughly two months, Antler made its first investment in Staff Inc.; it made 13 investments in total in 2018. Staff Inc. subsequently received seed funding from Golden Gate Ventures. The basic machine was already visible: find people → create talent density → form teams → validate problems and business models → make an Investment Committee decision → connect companies to outside venture capital. The Residency is what makes Antler resemble an accelerator, but economically it is better understood as investment sourcing and diligence infrastructure. Forbes described the 2022 historical model as typically lasting six to twelve weeks, during which participants searched for co-founders, formed business models and competed for Antler investment. Typical checks at the time were roughly $100,000–$200,000 for about 10% ownership, while participants could receive living stipends of up to $2,500. The model has since evolved. Antler’s current Fact Sheet describes a much shorter intensive residency for founders at the very beginning, while founders who already have a team and idea can use a fast-track process leading to an Investment Committee within weeks. Terms also differ by geography. The 2025 Nordic Fund II announcement, for example, described a two-tranche structure with roughly $100,000 at inception and another approximately $200,000 linked to the next financing round. Consequently, historic “10% for $100k–$200k” terms should not be treated as a universal or permanent Antler standard. Antler’s most powerful scaling advantage is an enormous founder funnel. In 2025 Magnus said Antler had already run more than 200 residencies and was receiving over 160,000 founder applications per year, roughly triple the level recorded in 2021. Yet high investment volume does not mean universal acceptance. Magnus said Antler invested in slightly more than four founders per thousand applications in 2021, falling to roughly 2.5 per thousand in the most recent annual period he discussed. At the same time, the Wall Street Journal reported that actual portfolio investments rose from roughly 100 in 2020 to more than 400 in 2025. Antler therefore achieved two apparently contradictory outcomes: absolute investment volume exploded, while the applicant pool grew even faster and overall selectivity increased. Antler Hub is an easily overlooked but strategically important operating asset. Antler began building proprietary founder-facing software early, launching the first version of Antler Hub in January 2020 and subsequently adding an internal management and reporting system initially called Fusion. Those tools were eventually combined into the current proprietary Antler Hub platform. The system connects founders, staff and external advisors and standardizes sourcing, selection, investment processes, research and reporting. Antler’s thesis is that each cohort and each investment creates more evidence about the attributes associated with successful founders and companies. Its potential moat is therefore: large founder funnel × multi-market outcomes × standardized proprietary data × local partner judgment. The capital base evolved from friends and personal contacts into a highly institutional LP network. Forbes reported that early major backers included Schroders, the International Finance Corporation and Eduardo Saverin. By 2025, Magnus said 70% of Antler’s AUM was institutional, including 19 sovereign backers. Representative institutions he named included M&G, Schroders, Phoenix Group, Macquarie, New Mexico State Investment Council and EIFO. Nordic Fund II closed at roughly $100 million in 2025 with investors including Formue, Ferd Impact, SmartCap, EIFO and Finland’s Tesi. In January 2026 Antler announced another $510 million of capital commitments across funds during the preceding 12 months. Participating institutions included New Mexico State Investment Council, EIFO, Tesi, an Australian superannuation fund, U.S. healthcare systems, university endowments and family offices. Roughly half of that new capital was expected to support U.S. startups. Fund LPs must not be confused with shareholders in Antler’s management company. Schroders, IFC, pensions, sovereign institutions and similar organizations are publicly identified primarily as investors or LPs in Antler-managed funds. That does not automatically establish ownership of the Antler management company itself. The complete corporate cap table of Antler, Magnus’s exact ownership percentage, and the economic split among the five co-founders are not publicly confirmed in the materials reviewed. It is therefore invalid to estimate Magnus’s personal net worth by treating Antler’s $1 billion-plus AUM as his wealth. Antler’s business model is a vertically integrated capital-value chain, not primarily a business that sells entrepreneurship courses. The first layer is founder acquisition: global branding, professional networks, operator networks, scouting and Residency applications generate a large pool of potential founders. The second is selection: partner interaction, team formation, rapid product validation, observation of founder behavior and Investment Committee decisions reduce the pool to a small set of investable companies. The third is very-early equity acquisition. In 2022, a founding Antler investor told Forbes that Antler could often enter companies at pre-money valuations around the $1 million vicinity. Entering that early can create exceptional upside if a company later reaches Seed and Series A/B valuations, but necessarily comes with high failure risk. The fourth layer is follow-on concentration. By 2025 Magnus said Antler had invested more than $20 million in a single company across multiple rounds. The fifth is Antler Elevate, the $285 million vehicle launched in 2023 to extend investing into later stages. It can invest in breakout Antler companies and external category leaders and gives Antler the ability to remain involved around Series A through Series C. By 2025 Antler said Elevate had deployed more than $160 million into over 60 companies. Antler has therefore evolved from a Day Zero specialist into a multi-stage venture platform extending from founder formation into growth capital. Its asset base becomes clearer when financial/operating assets are separated from influence assets. Operating and financial assets include the Antler brand and management platform, the portfolio equity held through its funds, proprietary Antler Hub infrastructure and data, and follow-on vehicles such as Antler Elevate. LP capital and portfolio holdings, however, belong within the fund structure and should not be treated as Magnus’s personal assets. Its harder-to-replicate influence assets include a founder funnel exceeding 160,000 annual applicants, an approximately 30-city local network, more than 70 Partners, Harvard/McKinsey/Rocket/startup alumni relationships, co-investor relationships and institutional LP credibility. Antler’s most valuable asset is therefore not a physical property or a media empire. It is a global matching network between entrepreneurial talent and capital. Antler deliberately treats other VC firms as partners rather than purely as competitors. Magnus’s 2025 retrospective says Antler has always tried to work with early- and later-stage investors, bringing outside VCs and angels into its ecosystem to mentor founders and ultimately become co-investors. This is economically rational. Antler’s early equity becomes far more valuable when portfolio companies attract subsequent institutional rounds. Helping another VC discover an Antler company is therefore not necessarily giving away value; it can be essential to increasing the survival probability and value of Antler’s own early stake. Its Advisory Board is another form of influence capital and is heavily weighted toward global finance, technology and policy elites. Antler’s current Fact Sheet names former McKinsey senior leader Tore Myrholt, former Goldman Sachs Asset Management Chairman Sheila Patel, former U.S. Treasury Secretary and Harvard President Lawrence Summers, former Facebook Chief Revenue Officer David Fischer, former Y Combinator Partner Aaron Harris, former Goldman Sachs Asia-Pacific CEO Ken Hitchner, and former Baillie Gifford investment leader James Anderson, among others. Such a network does not itself constitute investment performance, but it can materially support fundraising, institutional credibility, recruiting and portfolio introductions. The hardest external validation of Antler so far is that it has converted scale into top-tier industry activity rankings. PitchBook’s 2024 Global League Table Awards, announced in 2025, ranked Antler as the Most Active VC Firm globally, and also first in the Rest of World category. PitchBook’s methodology counts publicly disclosed transactions, so this is an activity ranking rather than a return ranking. Dealroom separately found that Antler made more AI investments than any other investor in 2024, ahead of a16z, General Catalyst, Sequoia and Khosla Ventures. These rankings matter because they demonstrate that Magnus’s once-radical ambition—to discover Day Zero founders systematically at global scale—has at least succeeded in terms of coverage and investment-production capacity. Airalo is one of the clearest current demonstrations that Antler’s early-stage model can produce a unicorn. In 2025, eSIM company Airalo raised approximately $220 million at a valuation above $1 billion. The round was led by CVC, with Peak XV and Antler Elevate among the participants. The strategic significance is not simply that Antler can list a unicorn. Airalo illustrates the intended lifecycle model: Antler can enter early and later use Elevate to deploy additional capital as a company matures. Lovable may be an even more important example of Antler’s exposure to the AI cycle. Antler’s January 2026 announcement said AI application company Lovable reached a valuation of roughly $1.8 billion after a $200 million Series A around eight months after launch, and was subsequently valued at $6.6 billion roughly two years after launch. Combined with Dealroom’s ranking of Antler as the most active AI investor of 2024, this suggests a potentially favorable structural interaction between AI and Antler’s model: as AI reduces the cost of building early software products, the number of potentially viable founder teams grows—and Antler was already designed to process an unusually large early-founder funnel. That is still an investment thesis, however, not proof of final fund returns. “Most active” must be separated from “best-performing.” PitchBook’s award is fundamentally a deal-activity ranking, not a fund-return ranking. Public Antler materials provide extensive data on deal counts, AUM, fundraising, unicorn valuations and subsequent financings, but do not provide a standardized platform-wide set of comparable net IRR, DPI and TVPI figures. Three common conclusions would therefore be incorrect: $1 billion of AUM does not mean Antler has earned $1 billion. A portfolio company valued at $6.6 billion does not mean Antler has realized an equivalent cash gain. Being the world’s most active VC does not mean being the world’s highest-return VC. The central historical criticism of Antler is whether extremely broad, high-frequency investing creates a statistical and data advantage—or dilutes investment judgment. In 2022 Forbes noted how crowded angel and pre-seed investing had become and argued that Antler still had to differentiate itself. At that time, Forbes reported that roughly one in eight Antler investments had failed and that the portfolio had not yet produced a unicorn or major exit. Airalo and Lovable have since answered part of that criticism. But the underlying issue remains relevant: the Wall Street Journal described Antler as intensifying a high-volume strategy, with more than 400 investments in 2025 versus roughly 100 in 2020. The question is therefore no longer whether Antler can find good companies—it clearly can. The harder question is whether the eventual winners across a portfolio of more than a thousand startups will compensate for failures, operating complexity and capital costs strongly enough to generate elite VC-fund returns. Its exit record remains much younger than its investment-count record. As of May 2026, Fortune reported that no Antler portfolio company had yet completed an IPO. That does not by itself imply failure: Antler only began investing in 2018, so much of the portfolio remains relatively young. But the standard by which Antler should now be judged will gradually shift from “How many companies did you fund?” toward: How many produce major exits? What does mature-fund DPI look like? What are the realized returns of the earliest fund vintages? Does high-frequency investing produce enough exceptional winners to transform whole-fund results? The Founder Residency also contains a structural tension between democratizing entrepreneurship and the economic cost of becoming a founder. Forbes’s 2022 description required participants to devote several weeks and, in many cases, leave existing employment while receiving living support of only up to roughly $2,500; participation did not guarantee investment. There is therefore an unavoidable tension between saying exceptional founders can come from any background and asking potential founders to absorb the opportunity cost of full-time entrepreneurial experimentation. Antler’s later move toward different local Residency formats and fast-track processes suggests that the model itself continues to be optimized. Magnus’s Rocket Internet background carries its own historical controversy, but this should be distinguished from personal legal or ethical scandal. Forbes notes that Rocket Internet has long been controversial in venture circles for rapidly reproducing business models proven in one market in another geography. Zalora emerged within that broader Rocket ecosystem. The main public criticism surrounding Magnus today is therefore not centered on personal legal or ethical cases. It is concentrated on whether Antler’s high investment frequency can generate superior returns, whether early-stage terms are attractive to founders, the opportunity cost of Residency participation, and whether global scale can be maintained without weakening judgment quality. The 2023 launch of Elevate was a crucial strategic pivot because discovering winners is not enough; Antler must retain meaningful exposure to them. One structural weakness of a Day Zero model is dilution. If the first check is small and a successful company raises large later rounds, the early fund’s percentage ownership can fall rapidly. The $285 million Antler Elevate vehicle was designed to address that problem by concentrating more capital in companies that demonstrate product-market fit and strong growth. That changes Antler’s economics. Large outcomes no longer depend only on “Did we discover the founder first?” but increasingly on “Once we discover a winner, can we continue owning enough of it?” The United States became a significant new strategic focus in 2025–2026. Antler opened a San Francisco office and Residency in 2025. In January 2026, it said roughly 50% of the $510 million in newly committed capital was expected to go toward U.S.-based startups. This does not necessarily contradict its original non-Silicon-Valley positioning. A mature Antler can pursue two strategies simultaneously: develop local informational advantages in under-covered ecosystems; and establish a direct presence in the world’s deepest technology-capital and AI-talent market. Magnus nevertheless continues to anchor Antler around the belief that innovation is geographically decentralized. Antler’s model uses on-the-ground teams because local investors are more likely to know the strongest founders before those founders become globally visible. Its own 2025 retrospective explicitly describes the combination of deep local support and global institutional reach as a central competitive advantage. His 2024 TechCrunch interview illustrates how strongly that worldview remains tied to Asia, including Southeast Asia’s digitization, India’s digital infrastructure, Japan’s ecosystem-building push and Korea’s industrial and consumer-technology capabilities. Antler’s future direction is increasingly clear: larger funds, deeper follow-on capacity and stronger technology infrastructure rather than simply opening endless conventional accelerators. Magnus’s 2025 roadmap included larger funds, a hub-and-spoke geographic model, more selective new-market entry, additional follow-on-capital mechanisms, and continued investment in technology and data systems. One specific mechanism is ARC—Agreement for Rolling Capital—designed to commit additional future capital beyond Antler’s initial investment. The institutional evolution can therefore be summarized as: “entrepreneur school + first check” becoming “global founder-discovery infrastructure + multi-stage capital platform.” A compressed timeline makes Magnus’s changing role especially clear. Around 1997–1999: UWC Atlantic College, his first sustained exposure to an international high-talent educational environment. Thereafter: roughly two years in the Norwegian Naval Special Forces, shaping his long-term ideas around endurance, rigor and purpose. 2003–2007: Harvard economics, during the rise of Facebook and social Internet platforms; long-term relationships such as Eduardo Saverin emerge. 2007: development work in northern Zambia. Approximately 2007–2013: McKinsey, adding technology-industry exposure and structured strategy training to his global network. Around 2012–2013: transition into Zalora in Singapore; public accounts differ on the exact transition year. His identity shifts from consultant to operator and co-founder. Thereafter: COO of Global Fashion Group, learning to operate across 26 markets. 2017: Antler created; Magnus moves from building one company to designing a system that repeatedly finds people capable of building companies. 2018: first Singapore Residency, 62 founders and 13 initial portfolio investments. 2019: Stockholm becomes Antler’s first office outside Singapore, beginning true geographic replication. 2020: first Antler Hub launched; investment activity reaches roughly the hundred-company-per-year scale. End of 2021: AUM slightly above $300 million. 2023: $285 million Antler Elevate closed, extending the platform into growth investing. 2024: PitchBook ranks Antler the world’s most active VC; Dealroom later identifies it as the year’s most active AI investor. 2025: more than 400 investments; Airalo and Lovable become headline unicorns; San Francisco launches; Nordic Fund II closes at $100 million. Early 2026: Antler announces $510 million in additional commitments raised during the previous 12 months and says its global portfolio exceeds 1,800 companies. Magnus’s present-day position is more complex than “entrepreneur who became a VC.” His defining skill is not traditional sector-specific stock picking, nor has he primarily built influence through books, media content or a personal publishing empire. A more accurate description is that he is a systems designer and resource orchestrator for entrepreneurial talent and capital. His accumulated resources form a sequence: rural Norway and UWC contributed a global-opportunity worldview; special forces contributed his resilience and execution framework; Harvard contributed a technology-era network; McKinsey contributed structured thinking and institutional relationships; Zalora/GFG contributed emerging-market operating experience; Antler productized all of the above. Antler’s most distinctive contribution is arguably not to a single industry, but to the question of when venture capital should enter the entrepreneurial process. Traditional VC generally prefers a founding team, a product and some evidence of traction before investing. Antler tries to move the investor one stage earlier: evaluate the person first, then help the company form. Its current Fact Sheet explicitly says it can invest when founders are still pre-team or pre-idea. It attempts to industrialize what was historically an accidental process—two talented people happen to meet, decide to build something, and happen to encounter an angel investor—into: sourcing → matching → formation → validation → investment → follow-on. That is Antler’s most original organizational contribution. Whether Antler ultimately becomes a Sequoia- or YC-caliber enduring institution will be determined by its next phase rather than by scale alone. Scale has been demonstrated: more than 1,800 investments, a footprint of roughly 30 cities, more than $1 billion of AUM and the No. 1 global VC activity ranking for 2024. Breakout-company discovery has also been demonstrated: Airalo and Lovable mean the idea that Day Zero selection can find major winners is no longer purely theoretical. Institutional fundraising has been demonstrated as well: 70% institutional AUM, 19 sovereign backers and another $510 million in recent commitments show that sophisticated LPs are willing to allocate meaningful capital to the platform. What remains to be fully demonstrated is mature realized fund performance. As Antler’s earliest portfolios reach exit age, DPI, IRR, major IPO/M&A outcomes and realized carry will determine whether Antler is merely one of the world’s most powerful founder-finding machines—or also one of its most successful venture franchises. The single clearest way to understand Magnus and Antler is this: his most important innovation was not picking one specific winning startup, but attempting to industrialize and globally reproduce the opportunity that repeatedly changed his own life—entering high-talent environments, meeting the right people and receiving access to the right resources. A person who began on a husky farm in rural Norway and entered progressively stronger networks through UWC, Harvard, McKinsey and Zalora ultimately created an institution designed to identify people who have not yet entered those networks, bring them inside, and acquire an equity stake in the value they subsequently create. Commercially, Antler is a founder-discovery + equity-acquisition + follow-on-concentration machine. As an influence structure, it is a global entrepreneurial-talent network. And Magnus himself is best understood not simply as a fund manager or founder, but as the chief architect, fundraiser, public face and long-term organizer of Antler’s global founder-capital system.
GMGN.AI: The Rise from Grassroots Meme Tool to Leading "Modular CEX" and In-depth Analysis of Its Founder
1. Overview: What is GMGN.AI and Who Leads It (1) Platform Positioning GMGN.AI is essentially a "multi-chain Meme trading terminal + data tool". Its product forms include a web platform, an app, and a Telegram bot. GMGN is not a traditional centralized exchange but rather an upper-layer tool that provides data analysis, trading routing, and automated execution services around decentralized exchanges like Pump.fun and Raydium. GMGN explicitly states in its Google Play disclosure that the app is not an exchange but a platform that combines data analysis, trading access, self-custody wallets, and custodial tools. (2) Founders and Key Figures GMGN has at least three identifiable core figures: Haze, known as "Chicken Brother" in the Chinese community, is a co-founder of GMGN and the most recognizable representative. He is primarily responsible for sales, branding, community operations, and external narratives, regarded by some community members as the soul of GMGN. Jerry Ma is the founder and CEO of GMGN, with a background in tool product development, having participated in the creation of NFTTrack. He is mainly responsible for product coordination, company management, and product development direction, serving as the core leader at the company level. Arthur's X account is @arthur_gmgn, and he is publicly identified as a co-founder of GMGN.AI, involved in team operations and external communications, but there is limited public information about him, mostly remaining at the social media identity level. (3) Product and Team Profile The GMGN team mainly consists of product managers and engineers with long-term experience in tool product development, most members having backgrounds in large Web2 internet companies or tool-based products. The team size has expanded from about 9-10 people initially to over 30, with personnel mainly focused on product, technology, and infrastructure roles, without establishing a large business development or marketing team. The internal role division is relatively clear: Jerry, as the product head and founder, leads the product direction. Haze is responsible for sales, branding, user operations, and external markets, positioning himself on the front lines. Other core members mainly handle product development, data processing, trading infrastructure, and security systems. 2. Family Background and Growth Environment (1) Birth Time and Region Haze states he was born in August 1987, belonging to the typical "post-85" generation, with the zodiac sign of Leo, and grew up in Hunan, China. He emphasizes his ordinary background, with no family capital or technical genius background. Currently, he has lived in Singapore for about four years, using it as a base to observe the global cryptocurrency market and operate GMGN. (2) Family Class and Parent Information Haze repeatedly emphasizes in interviews that he did not have grand dreams early on, with the most direct goal being to make money. He has stated that he previously worked in a very small place, following a typical ordinary career path. These statements point to a middle to lower-class workplace background in ordinary cities or counties in China, rather than a wealthy family, elite family within the system, or a family with abundant entrepreneurial resources. There is limited public information about his parents' professions, family asset status, and other family members. (3) Personality and Interests Public reports focus more on Haze's personality and interests after adulthood. His MBTI test result is INTJ-A. He considers himself a relatively introverted person but has to play a high-exposure, expressive extroverted marketing role on social media due to work needs. He enjoys rap music, including Eminem, Kendrick Lamar, Trap, and Memphis Rap. This cultural preference allows him to resonate more easily with the young Meme player community and reinforces his interest in sharpness, confrontation, and PVP game culture. He summarizes his personality as: generally introverted, inclined towards rational analysis and strategic thinking, but daring to bet on clear trends in product and market choices. This personality later directly manifested in key decisions like All in Meme and All in Solana. 3. Educational Background and Sources of Thought (1) Formal Education Haze clearly states that he did not attend university and does not come from a prestigious school or professional background. He first entered the workforce and later transitioned into the cryptocurrency industry due to Bitcoin news, self-learning quantitative trading, blockchain, and market knowledge. This experience is significantly different from the paths of many Web3 entrepreneurs who come from prestigious technical schools or have studied abroad, resembling the development path of frontline traders and practical tool developers. (2) Self-Learning Path and Knowledge Structure In 2018, Haze joined a quantitative company engaged in grid trading. He wrote an early systematic tutorial on grid trading on a blogging platform, gaining initial reputation in the quantitative trading circle. From public interviews, he is clearly influenced by the following categories of thoughts and texts: "The Art of War": Used to think about how to gain advantages through unconventional means. He views the market as a battlefield, understanding product iteration and trading games as military processes. "Tao Te Ching": The GMGN team has organized readings and sharing of the "Tao Te Ching", attempting to constrain the team in a more restrained and long-term manner, maintaining a certain moderation in the extremely PVP Meme market. Data transparency and self-correction: Haze emphasizes that one of the advantages of blockchain is its strong self-cleaning ability. Public data can break traditional venture capital narratives and help the market quickly identify errors and bubbles. This concept directly drives GMGN towards the product direction of "data retail + decision signalization". (3) Other Founders' Educational Background Jerry Ma is described as having about 20 years of experience in tool product development, mainly focused on internet tools and data products. Public reports do not disclose his specific educational background or school information. Arthur's educational experience and growth background have limited public information. 4. Early Work Experience: From Ordinary Employee to Quantitative Trader (1) Before Entering the Crypto Industry Before encountering Bitcoin, Haze states he only worked in a small place. His work status was to complete tasks and receive a salary, with no clear long-term career plan or entrepreneurial ideals. This phase is closer to the workplace life of ordinary young people in China's third and fourth-tier cities. (2) 2017-2018: Entering the Crypto Industry At the end of 2017, he accidentally saw Bitcoin-related reports in the news and intuitively bought a small amount of Bitcoin, viewing it as a worthwhile opportunity to try. This was his first exposure to crypto assets. In 2018, he officially joined a quantitative company engaged in grid trading, systematically learning blockchain and quantitative trading strategies from scratch. He quickly established a certain reputation in niche circles by writing tutorials and creating content. This phase laid the foundation for his later "trading + tool" capability combination. (3) Pionex, Pai Network, and BitUniverse Phase Haze subsequently entered the Pionex and Pai Network ecosystem, participating in the early construction of the Pai exchange and later becoming the head of BitUniverse in China. He was responsible for user growth, operations, customer service, and product-related work. He described himself as having done almost everything except coding. This experience deepened his understanding of centralized exchanges, quantitative tools, user growth, and trading product business models. This also explains why the market once mistakenly believed GMGN was a project incubated by Pai Network and why he has a strong sensitivity and iterative ability regarding automated trading tools. 5. Entrepreneurship and Project Experience: From NFTTrack to GMGN (1) Jerry Ma's Tool Product Path Before GMGN, Jerry Ma participated in creating NFTTrack. NFTTrack is a tool product that tracks blue-chip NFT trading dynamics, primarily serving professional NFT traders. The core logic of NFTTrack is "data converted into decisions", helping users discover trading opportunities through data dashboards, address tracking, and blue-chip asset trading monitoring. This experience was later transferred to GMGN, with the product focus shifting from NFTs to Meme Coins and multi-chain crypto assets. (2) Haze's Podcast, Content, and Community Operations Haze has long been active in podcasts, X Space, and Chinese crypto communities, participating in interviews and programs with organizations like Wu Says Blockchain and 7UpDAO. He publicly shared his experience of losing about $350,000 trading Meme assets, calling himself a "dog fool" and using this failure as a cautionary tale. This narrative of public failure amplified GMGN's community topic. "From $0 to $10 Million Meme Operation Guide" is another content product attempt. GMGN collaborated with several KOLs to systematically organize Meme trading methods, reinforcing its knowledge authority in the Meme scene while bringing in new users for GMGN. (3) GMGN: From Internal Startup to Independent Brand From the existing narrative, GMGN initially resembled an internal startup or amateur project pushed by Haze and former colleagues, not a large project relying on venture capital funding from day one. In terms of initial team size, product development pace, and funding choices, GMGN significantly differs from typical Web3 venture capital projects. Haze has repeatedly emphasized that GMGN is an independent project. Although many team members previously worked at Pionex or BitUniverse, there is no equity or incubation relationship between GMGN and these companies. This independence has also become an important statement for GMGN in responding to external doubts about its capital background. 6. GMGN's Timeline and Key Milestones (1) 2023: From Ethereum Launch to Seeking Direction In June 2023, GMGN officially launched and began development. The product initially connected to Ethereum, with the core entry point being smart money tracking, resembling a more user-friendly on-chain data dashboard and trading entry. From September to October 2023, GMGN began formal operations. At that time, the Meme market had not fully exploded, and the team mainly conducted product experiments around high-profile Meme assets like MOG and HPOS on Ethereum. (2) Q1 2024: Strategic Shift to Solana From January to March 2024, Haze observed projects like Silly Dragon, Mobile, and BOME on Solana experiencing successive explosions. After investing funds to participate in trading, he judged that the Meme market's funds and users would gradually settle on Solana. Therefore, he pushed the team to strategically shift to Solana starting in March 2024. During the migration process, there was significant internal controversy. The main concern was that if it took two months to complete the Solana version's development while Solana's popularity had declined, the team would waste critical time. Ultimately, the judgment that "active on-chain hunters are migrating to Solana" outweighed the risk concerns, and the team completed the strategic migration. (3) Q2 to Q3 2024: Function Explosion and Natural Growth In April to May 2024, GMGN's Solana support features were fully launched. The data dashboard and smart money tracking features were frequently used, and the platform gradually accumulated its first batch of deep users. In July to August 2024, GMGN launched the "Wash Trading" and "Insider Wallet" features, automatically marking project wallets, distribution wallets, and suspected internal trading addresses. This feature was extremely popular among professional Meme traders and was seen as a true breakout moment for GMGN. At the same time, Pump.fun's popularity rapidly increased. GMGN built a complete on-chain trading closed loop around Pump.fun's internal market and Raydium's secondary liquidity. Features like wash trading identification, developer rug history, and real-time risk alerts gradually formed differentiated barriers. (4) Q4 2024: Trading Volume and Revenue Explosion On November 17, 2024, according to third-party Dune dashboard data, GMGN's daily trading volume reached approximately $72 million. The daily trading volume ranked second only to PHOTON and higher than BULLX. The corresponding daily fee income was approximately $710,000, with about 25,000 daily active users. In the same month, Binance listed Meme assets like ACT and Pnut, further driving user attention to on-chain trading. GMGN, as an on-chain asset scanning and trading terminal, clearly benefited from this trend. Haze views this stage as a key node for Meme and on-chain trading to fully enter the mass market. (5) 2025-2026: Consolidating Leading Position Between 2024 and 2025, GMGN was referred to by some media as a "leading project in the modular CEX track". This concept refers to GMGN providing a trading experience and functional closed loop similar to centralized exchanges on-chain, while still being based on DEX and on-chain matching. Third-party tool evaluations generally regard GMGN as one of the leading platforms in Solana Meme trading tools. Related evaluations focus on three main aspects: Copy Trading user experience. Data update speed. Wallet tracking depth. 7. Product Forms and Technical Positioning (1) Asset Issuance and Data New Retail Haze views Meme as the "new retail era of asset issuance". In his analogy, Pump.fun is like an asset supermarket, allowing anyone to issue a large number of tokens at low cost and conduct rapid rotations. GMGN is akin to a data new retail platform, packaging complex on-chain data into simple trading signals, such as: Wash trading alerts. Smart money buying. Multiple wallet purchases. KOL purchases. Changes in project holdings. Users can directly complete trades based on these signals. Essentially, users are purchasing not just data but also the execution capabilities corresponding to the data. Haze believes that traditional data platforms like Nansen and Dune are more like providing engines, where users still need to build their decision models. GMGN aims to become a "4S store", directly providing users with complete products suitable for specific markets, namely "signals + one-click trading", lowering the threshold for Meme players to use on-chain data. (2) Core Functions GMGN's core functions include: New Coin Monitoring: Aggregating issuance platforms and liquidity pools like Pump.fun, FourMeme, Raydium, and filtering new coins based on market cap, trading volume, and creation time. Real-time Rankings and Trends: Ranking based on the number of buy/sell orders, transaction volume, price changes, and holder growth, displaying capital flows and market heat. Chip and Wash Trading Analysis: Identifying internal wallet ratios, developer holding ratios, concentration of the top ten addresses, and bulk buy orders. Security Audit: Checking whether liquidity is locked or destroyed, whether there are Honeypots, whether permissions are relinquished, and whether minting is possible, with some capabilities relying on external security services like GoPlus and Honeypot.is. Smart Money Tracking and Copy Trade: Scoring based on the historical win rate and profit/loss of wallets, supporting one-click copying of buy/sell operations and position ratios, and providing automatic profit-taking, stop-loss, and trailing stop-loss functions. KOL and Wallet Monitoring: Maintaining a list of high-value wallets and KOL addresses, providing real-time alerts for buying, selling, increasing positions, and liquidating. Trading Execution Layer: Connecting trading routes and DEX across multiple blockchains, providing multi-wallet batch order placement, Anti-MEV, priority fees, and customizable tips. Data API and AI Access: Providing structured APIs for Token, Wallet, Market, Signal, and Execution, encapsulated as AI Skills for GPT, Claude, and other agents to call for on-chain queries and trading execution. This functionality has garnered additional attention for GMGN in the AI Agent trading narrative in 2025-2026. (3) Wallet and Security Model GMGN offers two main usage modes: The first is the browser, app, and plugin wallet mode. Users authorize using their wallets like Phantom and MetaMask, requiring user signatures for each operation. This mode is relatively secure but has a lower user experience and automation level. The second is the Telegram Bot mode. The platform server hosts user keys or controls trading permissions through MPC and other methods. This mode is more suitable for high-frequency trading and automated strategy users but imposes higher demands on platform security and internal risk control. Haze has cited the private key plaintext transmission and server attack incidents of competitor DEXX as negative examples, emphasizing that GMGN established an independent security team from the beginning, using encrypted storage and isolated architecture. However, he also acknowledges that no system can achieve 100% security, and what truly matters is the ability to quickly identify problems, control risks, and compensate users. 8. Business Model and Revenue Structure (1) Basic Fee Structure: 1% Trading Fee GMGN's official documentation shows that the platform charges a 1% fee on each buy and sell transaction, with no monthly or subscription fees. Users' actual costs also include on-chain gas fees and fees charged by issuance platforms like Pump.fun. Some third-party evaluations indicate that trading through referral links can yield certain fee discounts. GMGN has also set up a referral commission mechanism, allowing referrers to earn a portion of the trading fees from invited users. (2) Revenue Scale and Profitability On the peak date in November 2024, GMGN's daily trading volume was approximately $72 million. Based on the 1% trading fee estimate, the daily fee income was about $710,000. Relevant data mainly comes from third-party Dune dashboards and media references. This revenue level once placed GMGN on the list of high-fee crypto infrastructure projects. Some statistical platforms show that GMGN's cumulative fee income reached approximately $229 million, indicating strong cash flow capabilities during the prosperous period of the Meme market. (3) Cost Structure and Expanded Revenue Jerry stated in an interview that GMGN's data analysis and real-time storage costs are high, with monthly server costs nearing $100,000. In the future, GMGN may open up advertising space, signal promotion, and paid API as revenue sources for projects. However, currently, the platform still mainly relies on trading fees. GMGN also provides data API whitelists, referral bots, and trading group commissions as cooperation models, offering monetization interfaces for projects, trading communities, and bot developers. (4) Choice Not to Issue Tokens Haze has publicly stated multiple times that GMGN's primary revenue source is trading fees. The team has not yet decided whether to issue a platform token. The core concern is that once a token is issued, users may focus excessively on the token price, neglecting the product itself, leading to a shift in product stickiness and team focus. Based on his observations of competition among NFT trading platforms, he believes that a project's real decline does not depend on whether it issues tokens but on whether it begins to ignore user needs. Therefore, GMGN prioritizes product iteration over building a Token economic system. 9. Capital Relations and Organizational Structure (1) Legal Entity and Registration Location Google Play's developer information shows that GMGN's development entity is GMGN Labs Limited. Its registered address is Trinity Chambers, Road Town, British Virgin Islands. The disclosed contact number is a Singapore number, and it uses emails like gmgn.ai and gmgn.cc. This structure resembles the "offshore company + Singapore operating team" model adopted by many crypto projects. (2) External Investment and Financing Public databases show that GMGN has not disclosed any funding rounds or public lists of institutional or angel investors. Combined with the financing situations of other projects in the same track, and GMGN's long-term emphasis on not blindly financing during market peaks, it can be reasonably judged that GMGN does not have any prominent venture capital institutions or Web3 funds invested at least at the public level. The platform mainly relies on self-generated income for rolling development. There is limited public information on whether there are undisclosed personal financing or equity arrangements. (3) Relationships with Other Organizations Relationship with Pionex and Pai Network: Several members of the GMGN team previously worked at Pionex or BitUniverse. However, Haze explicitly denies that GMGN is an incubated or controlled project of Pai Network, defining it as an independent project. Relationship with 7UpDAO: 7UpDAO partner Nathan has had multiple interviews and discussions with Haze, with their relationship mainly focused on content cooperation, opinion dissemination, and community communication, with no clear capital binding found. Relationship with centralized exchanges: Trading platforms like Bitget, Gate, and Binance have reported on GMGN in content and education. Currently, there is no clear information indicating that GMGN has become an official embedded tool or strategic investment object of any centralized exchange. 10. Key Decisions and Turning Points (1) All in Crypto and Quantitative Trading Transitioning from an ordinary employee to crypto and quantitative trading was a significant turning point in Haze's life. At the end of 2017, he decided to buy Bitcoin. In 2018, he entered a quantitative trading company, self-learning and writing grid trading tutorials. Thus, he completed the leap from an ordinary employee lacking educational advantages to a tutorial author and industry participant in the quantitative trading circle. This phase laid the foundation for his later multiple identities, including trader, KOL, product consultant, and founder. (2) Abandoning BRC-20 and Inscription Direction In June 2023, when GMGN was just starting its project, the team discussed whether to develop a BRC-20 wallet and inscription tools. Haze believed that inscriptions did not solve practical problems, had poor trading experiences, and low asset issuance efficiency, and could gradually lose vitality like some NFT projects in the long term. Therefore, he firmly opposed betting on the inscription direction and pushed the team to shift towards Meme Coin tools. In hindsight, this decision helped GMGN avoid the short-lived competition of inscription tools and seize the growth opportunities of Pump.fun and the Solana Meme market. (3) Migration from Ethereum to Solana This was one of the biggest turning points in GMGN's product development process. Before the migration, the Meme market's popularity on Ethereum was limited, and the trading experience was constrained by gas costs and network performance. The migration decision mainly stemmed from Haze's actual trading experiences with Solana Meme projects like Silly Dragon, Mobile, and BOME. He judged that active on-chain hunters were gathering on Solana. After the migration, GMGN's user numbers, trading volume, daily active users, and transaction counts experienced exponential growth. This was a high-risk, high-reward blockchain-level strategic bet that had a decisive impact on GMGN's market position. (4) Launching Wash Trading and Insider Wallet Features After establishing a foothold on Solana, GMGN launched features for wash trading identification, developer rug history, and chip distribution. These features compressed the originally high-threshold data analysis work into several labels, colors, and risk alerts, significantly improving decision efficiency for Meme traders. This functionality directly drove GMGN's reputation growth and community dissemination. Subsequent features like red fish, suspected collusion addresses, and sniper developer tags also collectively formed the core selling points of GMGN's data retail platform. This model shifted the Meme tool industry from "subscription-based professional data platforms" to "pay-per-trade + tool commission". (5) Choosing Not to Issue Tokens and Not to Raise Funds In the second half of 2024, the Meme tool track rapidly heated up, with a large influx of venture capital. GMGN chose not to issue a platform token or raise funds publicly but to rely on fee income for rolling development. The team publicly opposes short-termism that only chases trends and quick arbitrage. This choice sacrificed token market narratives and capital speculation gains in the short term but maintained a relatively pure product positioning and reduced some regulatory pressure. This also allowed GMGN to maintain a certain differentiation among trading tools that issued a large number of tokens. 11. Representative Achievements and Industry Impact (1) Status of Meme Tools and Trading Terminals According to multiple media and third-party evaluations, GMGN is in a leading position in the Solana and multi-chain Meme tool field. Its daily trading volume once ranked second in the Meme tool market, only behind PHOTON. In some fee rankings, GMGN has consistently ranked high, with cumulative fee income exceeding $200 million. GMGN has also been referred to by some media as a leading project in the modular CEX track, alongside BullX and Photon, forming the top platforms in the Meme trading tool field. (2) Data Retail and PVP Culture Haze's concepts of "the new retail era of asset issuance" and "data retail platform" provide GMGN with a clear product narrative. These concepts have also influenced the design direction of subsequent tool products, shifting the industry from providing APIs and data tables to directly offering decision signals and one-click trading. Haze openly accepts the PVP characteristics of the Meme market, understanding it as a high-risk gambling market with relatively transparent data. He emphasizes data openness, rapid trial and error, and small-scale participation. This perspective has changed some traditional value investors' views on the Meme market and prompted some institutions and seasoned investors to attempt using tools to build systematic Meme trading strategies. (3) Promoting Data and Trading Integration The GMGN team emphasizes that trading is the endpoint of data analysis. Therefore, the platform not only provides data dashboards but also deeply integrates order placement, automated strategies, Anti-MEV, and anti-squeeze trading execution capabilities. This approach promotes traditional on-chain data platforms towards "data + trading integration" and also impacts the traditional product boundaries of centralized and decentralized exchanges. With the rise of AI Agents, GMGN has standardized its API as AI Skills, allowing large models to directly call GMGN for on-chain queries and trading execution. This has given it a first-mover advantage in the agent trading field. 12. Controversies, Risks, and External Criticism (1) "Dog Fool" and Trading Failures Haze once lost about $350,000 in a dedicated account for trading Meme assets, being referred to by some X users as "dog fool" or "Meme Fool". From a traditional finance perspective, this is considered a serious trading mistake. Haze instead leveraged this label, packaging it as paying tuition and gaining experience through actual losses. He also detailed this experience in a podcast, drawing traffic to GMGN. This willingness to publicly acknowledge failure has earned him some recognition in the Meme community, which is heavily PVP-oriented. (2) Custodial Risks of the Telegram Bot GMGN's Telegram Bot mode requires hosting user private keys or controlling user assets through MPC and other methods. This type of model carries systemic risks across the industry. If the server is attacked, keys are leaked, or internal personnel engage in malicious behavior, user funds could suffer significant losses. Haze criticized competitor DEXX for issues like plaintext transmission of private keys and server attacks during discussions of security incidents. He emphasized that GMGN uses encrypted storage and isolated architecture. However, these security claims are difficult to fully verify externally. Thus, custodial risks remain a key concern for rational users and security researchers. (3) User Experience and Withdrawal Controversies In historical comments on Google Play, some users reported the following issues: Abnormal SOL balance display. Delays in asset arrival. Complicated withdrawal processes. At times, it required confirmation through the Telegram Bot or manually, preventing immediate completion of all withdrawals. These issues raised some users' doubts about withdrawal difficulties. Developers responded that these issues have been fixed and that withdrawals can now be made at any time. However, these historical records may still be viewed as negative experiences by some users, reminding external observers of the complexity of its fund flow and product architecture. (4) Capital and Compliance Opacity Public databases show that GMGN has not disclosed any financing records or issued platform tokens. The company is registered in the British Virgin Islands, with its operational focus in Singapore. This combination of "offshore company + high crypto fee income" naturally raises institutional concerns about its equity structure, taxation, anti-money laundering, and user fund paths. Currently, there are no significant reports of GMGN facing major regulatory penalties or legal lawsuits. However, considering its fee scale and Telegram Bot custodial model, GMGN may still face considerable compliance pressure in the future against the backdrop of increasing global regulation. Risks are particularly concentrated in custodial wallets, automated trading, and referral commission businesses. 13. Current Identity, Influence, and Structural Position (1) Haze: From Quantitative Tutorial Author to Meme Era Narrator Currently, Haze holds multiple identities: Co-founder and external representative of GMGN. A representative quantitative trader and tool entrepreneur in the Chinese and Asian Meme field. Proposer and disseminator of concepts like "the new retail era of asset issuance" and "data retail platform". He continues to emphasize long-termism, sharpness, and self-iteration through Chinese podcasts, interviews, and long articles. He attempts to position GMGN between a casino trading tool and an infrastructure provider in the extremely PVP Meme market. (2) Jerry Ma: The Low-Key Product Founder Jerry Ma rarely appears frequently on Chinese social media. However, in several in-depth reports in both Chinese and English, he is described as having about 20 years of experience in tool products and having led NFTTrack as a product-oriented CEO. His product thinking is closer to traditional internet companies: Small steps and quick iterations. Rapid MVP validation. Data-driven focus. Cautious about valuation and expansion. He places GMGN's main metrics on the number of traders, transaction counts, server stability, and product iteration speed, rather than solely focusing on trading volume. This indicates he prefers to understand GMGN from a tool product perspective rather than viewing it purely as a financial platform. (3) GMGN's Position in the Crypto Market Structure From the perspective of on-chain trading structure, GMGN is roughly positioned as follows: Upstream: Connecting liquidity and trading routes from Pump.fun, Raydium, Jupiter, various Layer 2 networks, and Solana DEX. Midstream: Building smart money, wash trading, and strategy signal layers through data collection, cleaning, and analysis. Downstream: Serving PVP Meme players, Meme KOLs, on-chain quantitative traders, as well as some venture capital institutions and professional traders. GMGN provides integrated tools from asset discovery, data analysis, trading execution to post-trade review. In the context of Meme becoming an important narrative in this crypto cycle, GMGN is both a "water seller" collecting tool fees and, in a sense, a rule maker. Its wash trading and risk labels can influence users' judgments about projects, giving GMGN structural influence that transcends a single tool in the Meme ecosystem. (4) Future Directions and Potential Changes Haze has judged that 2024-2028 may be a relatively important development cycle for the crypto market. His judgment is based on factors such as public participation of American politicians and tech entrepreneurs in crypto assets, as well as the rapid growth of on-chain trading scale. Under this macro judgment, GMGN's strategies include: Not rushing to issue tokens or to raise funds at high valuations, but relying on fee income for continuous expansion. Continuing to follow on-chain trends. Currently, the core market is Meme, but it may also migrate to other new narratives with similar data structures and user needs. Emphasizing sharpness and long-termism, while maintaining high-frequency product iterations and avoiding getting lost in capital and short-term speculation. GMGN's position in the real world is quite unique. On one hand, it is a high-income tool company registered in the British Virgin Islands, primarily relying on trading fees, and has not issued tokens. On the other hand, it is deeply embedded in the highly PVP and speculative Meme market. It serves as both an on-chain trading infrastructure and a tool and "arms dealer" on the periphery of the speculative market. GMGN's long-term development will be closely related to the scale of on-chain trading, the lifecycle of the Meme market, wallet security capabilities, and the global regulatory environment.
OKX Onchain OS Launches Agent Payments Protocol to Enable Autonomous On-Chain Payments and Transactions by AI Agents
...I projects, including Paxos, Solana, Aptos, Pharos Network, Nansen, Sui, 0G Labs, HeySorin AI, RootData, Corbits, QuickNode, Otto AI, Ethy AI, AltLayer, pieverse, Optimism, Surf AI, KITE AI, Mesh, Sahara AI, Teneo Protoc...
California Governor Gavin Newsom Signs Package of Bills Banning Public Officials from Issuing Meme Coins
...issuer profited about $636 million. On-chain analytics firm Nansen reported as of June 30, 2026, there were 988,905 loss addresses with a total unrealized loss of $3.81 billion, consistent with the official figures. <...
Cosmos Hub Halts Chain to Recover 1.227 Million Stolen ATOM Transferred to Multisig
...q5cn7tdrwwzlwm0e44fgq07e43ner6vph64, with signers including Nansen, Keplr, Enigma, Silknodes, Kiln, and Polkachu, only modifying this account. Approximately 500,000 ATOM had already been exchanged for ETH via THORChain b...
GOLD-related addresses sold off 80% of their holdings, deleting posts signals unloading and netting
.... The official Trump meme coin $TRUMP is a different asset. Nansen statistics show that as of the end of June, about 989,000 buying addresses had lost a total of about $3.81 billion, with about two-thirds of buyers losin...