Trillion-dollar capital shift · DeepSeek first round · Anthropic trillion IPO · SpaceX buys Cursor · $830B compute capex
If you are an AI founder, tech investor, or enterprise technology decision-maker, June 2026 may be the highest-density capital month you have ever seen: DeepSeek's $7.4B first external round, SpaceX's $60B all-stock acquisition of Cursor, Anthropic's $965B valuation surpassing OpenAI, and global cloud AI capex revised to $830B. Each deal resets industry benchmarks. This article delivers a 2026 mega-deal master table, deep dives on DeepSeek, Anthropic, OpenAI, SpaceX, and Manus, compute arms race data, eight trend signals, and a six-step decision runbook so you can move from watching to betting with a clear framework.
2026 is widely recognized as the AI funding supercycle year. From DeepSeek's record-breaking first round to SpaceX's $75B IPO—the largest in history; from Anthropic's $965B valuation surpassing OpenAI to OpenAI's confidential IPO filing—the capital density, deal size, and strategic intent of the entire AI industry are being rewritten.
| Deal / Event | Amount | Date | Type | Key Takeaway |
|---|---|---|---|---|
| DeepSeek first round | ~$7.4B (~51B RMB) | Completed 6/16 | Funding | Largest single AI round in China; Tencent and CATL participate |
| Anthropic Series H | $65B | Completed 5/28 | Funding | $965B valuation, first to surpass OpenAI |
| OpenAI confidential IPO filing | — | 6/8 | IPO | Filed within a week of Anthropic |
| Anthropic confidential IPO filing | — | 6/1 | IPO | Expected listing October 2026 |
| SpaceX IPO | Raised $75B | Listed 6/12 | IPO | Largest IPO ever; $1.77T valuation |
| SpaceX acquires Cursor | $60B | Signed 6/16 | Acquisition | AI coding tool; all-stock transaction |
| Manus AI buyback | ~$2B | 6/18 | Buyback | Chinese investors repurchase from Meta after regulatory order |
| Baseten funding | $1.5B | 6/19 | Funding | Valuation jumped from $5B to $13B in five months |
| OpenAI 2025 spending | $34B | Disclosed 6/16 | Financials | Revenue $13B; spend 2.6x revenue |
Facing this capital wave, decision-makers often fall into these blind spots:
Confusing valuation with revenue: Anthropic at 20.5x PS, OpenAI at 65.5x PS, SpaceX at roughly 590x revenue—whether public markets can absorb these multiples is the biggest question of H2 2026.
Ignoring strategic capital logic: Tencent and CATL backing DeepSeek marks a shift from financial VC to industry-plus-strategy investing.
Underestimating geopolitics: The forced Manus unwind is the AI industry's first cross-border acquisition reversed by national regulators.
Training-only mindset: Baseten and other inference infra players are attracting capital as compute consumption shifts from training clusters to inference serving layers.
Stale industry boundaries: A space company buying an AI coding tool, a battery maker investing in a foundation model—the AI capital map has no fixed borders.
On June 16, 2026, DeepSeek officially closed its first external funding round, raising over $7.4B (~51B RMB) at a post-money valuation above $50B (~338B RMB). This is the largest single AI funding round in China's history.
| Element | Details |
|---|---|
| Founder co-investment | Liang Wenfeng personally invested ~$2.8B (20B RMB), the largest single check |
| Largest external investor | Tencent ~$1.4B (10B RMB) |
| Strategic capital | CATL ~$700M (5B RMB, including Puquan Capital) |
| Other investors | NetEase, JD.com, Monolith, IDG ~$420M each (3B RMB); Zhenfund, Shixiang ~$210M each (1.5B RMB); National AI Industry Investment Fund ~$140M (980M RMB) |
| Special structure | External capital flows into a limited partnership managed by Liang Wenfeng. Investors have no voting rights but receive priority financial information and follow-on rights |
| Lock-up | Five years; shares cannot be transferred |
| LP due diligence | DeepSeek team required verification of ultimate LP identities behind all contributing funds |
| Single exception | National AI Industry Investment Fund invests directly in DeepSeek's operating entity with voting rights and no lock-up |
Tencent faces entry anxiety in AI. Its in-house Hunyuan Hy3 preview performs well, but DeepSeek's technical lead makes a $1.4B external AI entry cheaper than catching up internally. WeChat, ads, games, cloud, and enterprise services all need stronger AI capability. DeepSeek can plug into Tencent's product ecosystem. Tencent has already backed Zhipu, MiniMax, Moonshot, StepFun, and Baichuan.
| Date | Valuation |
|---|---|
| Early April 2026 (secondary market) | ~$10B |
| Round launch | Target $35–59B (350–400B RMB) |
| After close | Above $50B (338B+ RMB) |
Valuation multiplied roughly fivefold in two months because DeepSeek V4 open-sourced and gained global recognition; Tencent and CATL provided strategic validation; AI sector multiples expanded broadly; and the founder's $2.8B co-investment signaled strong control confidence.
On May 28, 2026, Anthropic closed a $65B Series H at a $965B post-money valuation, surpassing OpenAI (then ~$852B) for the first time.
| Metric | Anthropic | OpenAI |
|---|---|---|
| Latest valuation | $965B | ~$852B |
| Annualized revenue | ~$47B (May 2026) | ~$25B |
| 2024 loss | ~$56B | Not disclosed |
| Profitability outlook | 2028 free cash flow $17B | Profitable by 2030 |
| Enterprise revenue share | ~80% | Not disclosed |
| $1M+ customers | 1,000+ (April 2026) | Not disclosed |
Anthropic's growth drivers: Claude Opus 4.8 leads ScienceQA at 76.4; Claude Code runs ~$6.3B ARR with a 54% share of AI coding agents; Constitutional AI builds enterprise trust; 80% of revenue comes from enterprise customers including eight Fortune 10 companies. IPO timeline: confidential S-1 filed June 1; expected listing October 2026; target first-day market cap $1.10–1.25T; expected raise $25–35B.
On June 16, 2026, the Financial Times reported OpenAI's 2025 full-year spend at $34B against revenue of only $13B—$2.60 spent for every $1 earned. Breakdown: R&D ~$19B, sales and marketing ~$6B, infrastructure and headcount for the remainder. Revenue highlights: beat internal $10B target, nearly 1B global users, ChatGPT share dipped below 50% for the first time but remains number one. IPO: confidential S-1 filed June 8; expected Q1 2027 listing; target first-day market cap ~$1.08T; raised $122B in March at $852B valuation.
| Dimension | Anthropic | OpenAI |
|---|---|---|
| Valuation | $965B | $852B |
| Annualized revenue | ~$47B | ~$25B |
| IPO timing | October 2026 (expected) | Q1 2027 (expected) |
| First-day market cap forecast | $1.10–1.25T | ~$1.08T |
| Profitability timeline | 2028 FCF | 2030 |
| Core strengths | Enterprise trust, safety alignment, Claude Code | User scale, ecosystem depth, GPT-5.5 |
| Core risks | Fable 5 export control incident | Persistent losses, governance |
Per DealRoom and other trackers, 2026 AI IPO proceeds could exceed $3.12T—potentially more than all U.S. IPOs since 2022 combined, the most concentrated thematic IPO wave since the late-1990s tech boom.
| Company | Target Valuation | Expected Timing | Sector |
|---|---|---|---|
| SpaceX | $1.5T | H2 2026 | Space / AI compute |
| OpenAI | ~$1T | Q1 2027 | AI / foundation models |
| Anthropic | $1.10–1.25T | October 2026 | AI / foundation models |
| Databricks | $134B | Q3 2026 | AI / data |
| Canva | $42B | Q3 2026 | Design / SaaS |
| Revolut | $75B | Q4 2026 | Fintech |
| Kraken | $20B | Q3 2026 | Crypto |
| Discord | $15B | Q2 2026 | Social / gaming |
On June 16, 2026—just four days after SpaceX's $75B IPO—the company announced a $60B all-stock acquisition of Cursor parent Anysphere, expected to close in Q3 2026. Cursor's ARR exceeded $4B by early June, making it one of the fastest-growing AI developer tools. Real coding data strengthens xAI Grok training and puts SpaceX in direct competition with Anthropic and OpenAI on AI coding.
| Dimension | Details |
|---|---|
| SpaceX valuation | Surpassed Amazon at $2.7T, fifth globally |
| xAI strategy | Acquire high-quality coding data to accelerate Grok in programming |
| AI coding landscape | Claude Code, GitHub Copilot, OpenAI Codex, and Cursor reshape the category |
| Signal | A non-AI company (aerospace) enters the arms race via acquisition; industry lines blur completely |
SpaceX's IPO prospectus reveals a broader AI strategy: Starlink is not just broadband but potential AI data center infrastructure; Starship could support orbital data centers and large satellite deployments; a $6.3B Reflection AI compute agreement is signed ($150M/month); committed revenue exceeds $80B, including $1.25B/month from Anthropic and $920M/month from Google. SpaceX is now Starlink plus rockets plus AI infrastructure plus Mars vision stacked together.
| Date | Event |
|---|---|
| December 2025 | Meta acquires Manus for ~$2B (Singapore-registered, China-founded team) |
| April 27, 2026 | China's NDRC orders Meta to unwind the acquisition |
| May 2026 | Meta begins internal data isolation, stops sharing data with Manus |
| June 18, 2026 | Early Chinese investors (HSG, ZhenFund, Tencent) plan to repurchase at the original $2B price from Meta |
| June 2026 | Manus annualized revenue rose from ~$100M at acquisition to $400–500M |
This is the AI industry's first cross-border acquisition forced to unwind by national regulatory intervention. Implications: AI assets are strategically sensitive; deal structures must pre-assess regulatory risk; industry capital plus geopolitics is a new variable; Manus is considering restructuring as a China joint venture targeting a Hong Kong listing. HSG and ZhenFund may raise new capital to buy Meta's stake; Benchmark opted out of the buyback.
Inference infrastructure player Baseten grew from a $5B to $13B valuation in five months on a latest $1.5B round—positioned at the enterprise AI model inference layer, filling gaps OpenAI and Anthropic do not directly serve.
| Company | Amount | Sector | Highlight |
|---|---|---|---|
| Sand.ai | >$100M (two rounds) | Video generation AI | Magi-1 Physics IQ ranks first; VidMuse hit $10M ARR in three months |
| Zhipu AI | Undisclosed | Foundation models | GLM-5.2 open source leads; coding beats GPT-5.5 |
| MiniMax | Undisclosed | Foundation models | M3 MoE architecture; only 23B active parameters |
| Moonshot (Kimi) | Undisclosed | Foundation models | K2.7 Code released; ARR passed $100M |
| Enflame Technology | IPO approved | AI chips | STAR Market IPO cleared |
| MicroNano Core | >$14M Series B | Compute-in-memory AI chips | Rising player in compute-in-memory |
TrendForce's May 2026 forecast: global Top 9 cloud vendors' combined 2026 capex was revised to about $830B, with YoY growth raised from 61% to 79%.
| Vendor | 2026 Capex | YoY Growth | Notes |
|---|---|---|---|
| Amazon / AWS | ~$200B | — | Reaffirmed guidance |
| Microsoft | ~$190B | ~130% | Includes $25B from component price increases |
| Google / Alphabet | $180–190B | >100% | Raised sharply from initial $17.5–18.5B guidance |
| Meta | $125–145B | ~85% | Raised from initial $11.5–13.5B guidance |
| Oracle | Plans to raise $50B | — | Expanding AI infrastructure |
| ByteDance | ~$200B | Raised 25% | Enters global AI spend top tier |
| Tencent | Q1 2026 capex $4.4B (31.9B RMB) | — | Continued AI investment ramp |
| Alibaba | Long-term commitment well above $380B | — | Multi-year pledge |
Key insight: the top five North American cloud vendors will spend roughly $545B on AI capex in 2026, about 75% of total capex. AI servers will exceed general-purpose servers in total power draw for the first time in 2026. Transformer and power distribution supply chains face longer lead times. North American data center vacancy fell to ~1.4% (JLL); pricing power is shifting from cyclical to structural.
| Source | Forecast | Time Horizon |
|---|---|---|
| McKinsey | $6.7T global data center construction cost; ~70% AI-related | Through 2030 |
| Morgan Stanley | $2.9T global AI infrastructure investment | Through 2028 |
| NVIDIA Jensen Huang | $3–4T total AI infrastructure spend; compute demand doubles every 100 days | Through 2030 |
AI IPO supercycle is here: 2026 AI IPO proceeds may exceed $3T; SpaceX, OpenAI, and Anthropic test public markets simultaneously.
Valuation vs profitability tension: Anthropic 20.5x PS, OpenAI 65.5x PS, SpaceX ~590x revenue.
Strategic capital dominates: Tencent and CATL in DeepSeek mark a shift from financial VC to industry-plus-strategy logic.
Geopolitics is the largest variable: Manus forced unwind shows AI assets are on strategic sensitivity lists.
Compute shifts from availability to affordability: Inference is the main consumption scenario; Baseten and peers attract capital accordingly.
Open-source commercialization paradox: DeepSeek V4 is MIT open source yet the round emphasizes commercialization—balancing community influence and revenue loops is the core question.
AI talent war intensifies: Google DeepMind lost Noam Shazeer (Transformer co-inventor) and Nobel laureate John Jumper to OpenAI and Anthropic within 48 hours.
From model company to compute company: Rockets are a small slice of SpaceX's $1.77T valuation; Starlink plus AI compute infrastructure carries the narrative.
Build a deal radar: Track DeepSeek, Anthropic, and OpenAI IPO windows plus SpaceX-Cursor closing progress. Avoid irreversible capital commitments during information vacuums.
Separate financial from strategic investing: Assess whether partners chase PS multiples or strategic assets like compute-power synergy, coding data, and enterprise trust.
Front-load cross-border compliance: Use the Manus case to embed regulatory unwind clauses and data isolation plans at term sheet stage.
Shift compute budget from training to inference: Following TrendForce's $830B capex trend, prioritize inference serving vendors (Baseten-class) and self-build cost curves.
Stress-test with PS multiples: Use Anthropic 20.5x, OpenAI 65.5x, and SpaceX 590x as upper-bound scenarios to test whether your AI product unit economics support the valuation narrative.
Align tech stack with capital cycles: The IPO supercycle will amplify model and API price volatility. Configure multi-vendor routing for production pipelines. For coding tool selection, see our AI coding assistants comparison.
2026 is not a year to watch from the sidelines—it is a year to place bets. Whether you are a founder, investor, or enterprise decision-maker, understanding this capital reset is how you find signal inside uncertainty.
Chasing AI capital headlines, many teams choose pure cloud APIs or self-owned GPU clusters. The former lacks buffer against export controls, quota throttling, and pricing swings. The latter carries hidden costs: hardware procurement, 24/7 ops, and Metal toolchain maintenance. For production environments that need stability for iOS CI/CD and AI Agent automation, VpsMesh Mac Mini cloud rental is usually the better fit—scale remote Mac nodes on demand, run Agent pipelines and Xcode builds in isolated environments, without buying hardware or running a data center. See Mac Mini M4 rental pricing and the cloud order page.
External capital flows into a limited partnership managed by founder Liang Wenfeng. Investors have no voting rights but receive information rights and follow-on priority. A five-year lock-up applies. The National AI Industry Investment Fund invests directly in DeepSeek's operating entity with voting rights and no lock-up. Liang Wenfeng co-invested ~$2.8B, the largest single check.
On May 28, 2026, Series H raised $65B at a $965B post-money valuation (OpenAI was then ~$852B). Annualized revenue reached ~$47B with 80% from enterprise customers. Claude Code runs ~$6.3B ARR with a 54% share of AI coding agents. Expected IPO October 2026; target first-day market cap $1.10–1.25T.
Cursor's ARR exceeded $4B by early June. Real coding data strengthens xAI Grok training. The $60B all-stock deal was announced four days after the IPO; closing expected Q3 2026. Claude Code, Copilot, Codex, and Cursor reshape the category. For tool selection, see our AI coding assistants comparison.
On April 27, 2026, China's NDRC ordered Meta to unwind the acquisition. On June 18, HSG, ZhenFund, Tencent, and other early investors planned to repurchase from Meta at the original $2B price. Annualized revenue rose from ~$100M to $400–500M. This is the AI industry's first cross-border deal forced to unwind by national regulators.
TrendForce forecasts combined capex for the global Top 9 cloud vendors at ~$830B in 2026, YoY 79%. McKinsey projects $6.7T in global data center construction through 2030, roughly 70% AI-driven. For stable local Agent and CI environments, see the help center for Mac Mini M4 cloud deployment.
Configure multi-model routing, watch API pricing after IPOs, and run Agent plus Xcode CI on remote Mac nodes to reduce single-cloud dependency. Review rental pricing and provision nodes on the order page.