Venture Capital
3.15M subscribers
1.9K photos
41 videos
1.95K links
The official channel of V3V Ventures. We share updates on our investments, portfolio companies, and fund activities.

Buy Ads: @strategy (this is our only account).
Download Telegram
🤔 Is AI a bubble? or the new railroads?

AI infrastructure spending has reached historic levels, with data centers alone driving 33% of US GDP growth in 2025. That’s $3T in the US and another $1T worldwide, numbers that recall past bubbles, but with key differences.

🖱 In scale: AI accounts for <1% of GDP, less than telecom (1.2%) and far below the railroads of the 1870s (4%), whose crash was brutal.
🖱 In capital intensity: GPUs depreciate faster than almost any asset (“chips depreciate like fish,” says Bill Gurley), leaving concrete and cooling as the only salvage value.
🖱 In revenue: AI revenues jumped from ~$0 five years ago to $60B today, doubling annually, unlike past bubbles, where growth slowed before the collapse.
🖱 In valuations: Nasdaq’s P/E peaked at ~72 in 2000; today it sits at ~32 high, but still tethered to earnings.
🖱 In funding: Unlike debt-fueled bubbles, today’s AI buildout is financed mostly from Big Tech free cash flow, not leverage.

The catch? Sustaining this boom will require another $2.9T. Giants can cover half, but outside capital must carry the rest.

AI today looks less like a bubble and more like an intensive investment boom, underpinned by real, accelerating revenue growth, a key distinction from railroads and dot-coms.


✔️Powered by Trade Watcher
Please open Telegram to view this post
VIEW IN TELEGRAM
Please open Telegram to view this post
VIEW IN TELEGRAM
8👍1
🚀 International Venture news for the past week

🖱 Healthtech consolidation accelerates: 2025 has already seen 102 healthtech M&A deals, nearly matching last year’s full count (112). The sector is rapidly consolidating as scale becomes survival. 🔗 WSJ

🖱 VCs use AI to source startups: San Francisco’s Touring Capital deployed its “R2-D2” algorithm to scout deals, closing a $330M fund for AI startups. AI is moving upstream into venture itself. 🔗 WSJ | Economic Times

🖱 Brazilian legal AI attracts US giants: Enter, a São Paulo startup automating legal workflows, raised $35M from Founders Fund and Sequoia at a $350M valuation. Clients include Airbnb and Nubank. 🔗 WSJ

🖱 Fewer deals, larger checks
2025 deal value up 42.8%, but deal count down 31.4%. The shift favors bigger rounds for fewer winners. 🔗 Axios

🖱 Pharma doubles down on venture: Sanofi Ventures boosted its capital to $1.4B, focusing on biotech and digital health bets. Big Pharma is increasingly playing as LP + operator. 🔗 MedCityNews

✔️Powered by Trade Watcher
Please open Telegram to view this post
VIEW IN TELEGRAM
111👍1
🔔 Trump’s DOE bans “climate change” and “green” from staff vocabulary

The U.S. Department of Energy’s Office of Energy Efficiency & Renewable Energy has circulated a memo forbidding staff from using terms like “climate change,” “green,” “decarbonization,” and “sustainability.” The directive comes as the Trump administration reframes federal energy policy.

🖱 The banned list also includes “emissions,” “carbon footprint,” and “transition,” signaling a retreat from standard climate terminology.
🖱 The change is led by a Trump-appointed official and reflects political positioning rather than scientific consensus.
🖱 Global clean energy investment, meanwhile, hit a record $386B in H1 2025, highlighting the gap between U.S. rhetoric and market momentum.
🖱 DOE staff now face constraints in reporting, grants, and external communication, raising concerns over transparency and accountability.

By policing language, the administration isn’t just shaping optics, it risks reshaping how U.S. energy policy is written, tracked, and enforced.


✔️Powered by Trade Watcher
Please open Telegram to view this post
VIEW IN TELEGRAM
6👍4🤔2😎1
💰 TikTok’s US sale looks like a steal

The Trump-approved deal values TikTok US at $14B, a fraction of what comparable platforms trade for. The app generates an estimated $10–20B annually in the US, yet sold for roughly 1× revenue.

🖱 By comparison: Snap trades at ~2.5× revenue, Reddit at ~25×, making TikTok’s price look like daylight robbery.
🖱 The Chinese side has not commented publicly, leaving questions about consent or coercion.
🖱 Trump claims the agreement is final, though details on buyer structure and governance remain vague.

If the valuation holds, this may go down as one of the most lopsided deals in tech history.

✔️Powered by Trade Watcher
Please open Telegram to view this post
VIEW IN TELEGRAM
8🤓2
🔔 How Larry Ellison Kept 40% of Oracle

Oracle’s founder still owns ~40% of his $300B company, a rare feat when most tech founders get diluted down to 5–15%. The reason? In 1977, no venture capitalist would fund software.

🖱 Late ’70s VC dollars flowed to hardware and biotech, Intel, Apple, Genentech all raised millions.
🖱 Ellison couldn’t even get a meeting. He and two co-founders pooled $2,000 and lived off consulting.
🖱 Their first break was a $50K CIA contract; nights were spent building their own database product.
🖱 Oracle launched in 1979 with “Version 2” (nobody trusted a Version 1). By 1984 revenue hit $70M; by 1987 it was the world’s largest database company.

By necessity, Ellison bootstrapped, making money from day one, avoiding dilution, and keeping control.

The irony: being ignored by VCs turned $2K of founder capital into one of the greatest fortunes in tech history.


✔️Powered by Trade Watcher
Please open Telegram to view this post
VIEW IN TELEGRAM
19🔥54
🎓 MBAs in Venture Capital

PitchBook tracked where VC firm employees earned their MBAs. The leaders look familiar, HBS, Stanford, Wharton, but unlike unicorn founder charts, the distribution is much smoother.

🖱 Top schools still show up, but the drop-off is gentle rather than steep.
🖱 Normalized by class size, the gap would shrink even further.
🖱 Takeaway: for breaking into VC, the brand of your MBA matters far less than in founder circles.

VC isn’t about heavy lifting, any solid MBA can open the door.

✔️Powered by Trade Watcher
Please open Telegram to view this post
VIEW IN TELEGRAM
5👏1
💡 Bill Gates on Why You Shouldn’t Compare Yourself

“Don’t compare yourself with anyone in this world. If you do so, you are insulting yourself.” ~ Bill Gates

In venture (and in life), comparison is a trap. Others’ wins often look effortless, but you rarely see the mistakes, resources, and timing behind them.

🔥 What matters:
🖱 Focus on your own path: test your ideas instead of copying others.
🖱 Learn from experience: treat others’ wins and failures as lessons, not standards.
🖱 Think like a portfolio investor: diversify approaches instead of chasing one “perfect” play.

In Business @ the Speed of Thought, Gates explains how systems thinking lets you make decisions without looking back at competitors, staying ahead by design, not by comparison.

A useful selection of Bill’s books:
1. Business @ the Speed of Thought (1999): how IT and “digital nervous systems” transform decision-making.
2. How to Avoid a Climate Disaster (2021): a roadmap to net zero with industry-specific solutions.
3. The Road Ahead (1995/96): forecasts of the digital revolution and its societal impact.
4. Showing Up for Life (2009): wisdom and life lessons from Gates’ father.


Venture is a marathon, not a sprint. Your victories are unique, your mistakes are priceless, don’t trade them away by measuring against others.

✔️Powered by Trade Watcher
Please open Telegram to view this post
VIEW IN TELEGRAM
20
🔔 Robinhood wants to let everyone invest in startups

Robinhood has filed to launch Ventures Fund I, a new vehicle that would give retail investors access to early-stage startups, something usually reserved for VCs, institutions, and the ultra-wealthy.

🖱 The SEC filing outlines a venture fund targeting sectors like AI, robotics, aerospace, and fintech.
🖱 Unlike traditional venture funds, this one is pitched as “open to all” regular Robinhood users could buy in.
🖱 Exact details on fees, minimums, and launch timing are still unclear.
🖱 It’s part of Robinhood’s push to expand beyond trading into broader wealth and investing products.

If it works, Robinhood could democratize venture investing, but it also means everyday investors taking on the same high-risk, high-reward bets as VCs.

The open question: will retail see this as empowerment, or as Robinhood pushing risky assets onto its users?


✔️Powered by Trade Watcher
Please open Telegram to view this post
VIEW IN TELEGRAM
11👀3
💰 EA Goes Private in a Record $55B Buyout

Electronic Arts is set to go private in the largest leveraged buyout ever, a $55 billion deal led by Saudi Arabia’s Public Investment Fund (PIF), Silver Lake, and Affinity Partners. Shareholders will receive $210 a share, a 25% premium.

🖱 The deal is financed with ~$36B in cash and ~$20B in debt, a massive leverage bet.
🖱 EA’s cash cow franchises like FIFA and Madden give investors confidence in steady recurring revenue.
🖱 Timing is key: the deal lands just before the launch of Battlefield 6, EA’s biggest new release in years.
🖱 Going private shields EA from quarterly earnings pressure, letting it refocus on long-term strategy.

This isn’t just about gaming, it’s about capital markets shifting. Big LBOs are back, sovereign wealth money is flowing into entertainment, and private ownership is reshaping the future of blockbuster gaming.


✔️Powered by Trade Watcher
Please open Telegram to view this post
VIEW IN TELEGRAM
15
💳 Ex-OpenAI & DeepMind Researchers Raise $300M to Automate Science

Periodic Labs, a new startup from AI veterans Liam Fedus (ex-OpenAI) and Ekin Dogus Cubuk (ex-DeepMind), just pulled in a $300M seed round to build autonomous labs powered by AI “scientists.”

🖱 Fedus helped lead large language model research at OpenAI; Cubuk pioneered AI-driven materials discovery at Google Brain/DeepMind.
🖱 Their vision: robotic labs where AI designs experiments, runs them, collects data, and iterates without humans in the loop.
🖱 First focus: superconductors, materials that could unlock breakthroughs in energy and computing.
🖱 The pitch: turn science into scalable software, cutting years of trial-and-error into rapid AI-driven discovery cycles.

If it works, Periodic Labs could redefine R&D itself, shifting science from slow human bottlenecks to autonomous discovery engines.


✔️Powered by Trade Watcher
Please open Telegram to view this post
VIEW IN TELEGRAM
5👍2
🚀 Musk becomes the first $500B man

Elon Musk just became the first person in history to see his net worth hit $500 billion, thanks to Tesla’s rebound and soaring valuations for SpaceX and xAI. By the market close, it dipped slightly to $499.1B, but the milestone was crossed.

🖱 Tesla remains the core driver: Musk still owns ~12% of the EV giant, whose stock has surged this year.
🖱 SpaceX and xAI have added billions in private-market value, boosting his fortune far beyond Tesla alone.
🖱 Tesla’s board has dangled a new $1T compensation package, tied to aggressive growth targets.
🖱 Much of the wealth isn’t liquid, it’s paper value tied to stock and private equity, making his net worth highly volatile.

For Musk, hitting half a trillion isn’t just a personal record, it’s a signal that the world’s richest person is building wealth on an unprecedented scale, faster than anyone before.


✔️Powered by Trade Watcher
Please open Telegram to view this post
VIEW IN TELEGRAM
22👏4
🤖 OpenAI crowned the first $500B startup

OpenAI’s secondary share sale has vaulted its valuation to $500 billion, making it the most valuable private company ever and a symbol of how far the AI boom has run. Unlike a traditional fundraise, the deal let insiders cash out while investors fought for a seat at the table.

🖱 Roughly $6.6B worth of employee and early-backer stock changed hands.
🖱 Heavyweights like SoftBank, Thrive, Dragoneer, MGX, and T. Rowe Price joined the cap table.
🖱 The valuation leap, from $300B just months ago, reflects both AI demand and scarcity of access.
🖱 Still, OpenAI is burning cash, betting scale and distribution will cement its lead over rivals.

The $500B milestone isn’t just about valuation, it’s proof that in AI, investors are willing to pay sovereign-nation prices for a private company.


✔️Powered by Trade Watcher
Please open Telegram to view this post
VIEW IN TELEGRAM
5🔥53
🔔 Durov opens Telegram’s AI lab in Kazakhstan

Pavel Durov has just launched a specialized AI laboratory in Kazakhstan, partnering Telegram with the country’s AI supercomputer cluster to extend AI access globally.

🖱 The lab is housed in the Alem.ai building as part of a joint project with Kazakhstan’s Ministry of AI and the national supercomputer cluster.
🖱 Durov says the goal is to let over a billion people use AI tools in a way that’s confidential, transparent, and efficient.
🖱 Telegram’s own mini-apps will be among the first to leverage the new tech.
🖱 The Kazakh supercomputer cluster is expected to be a major provider of compute power for the network.

This move signals Telegram’s pivot from messaging into AI infrastructure, using the state’s compute assets to deepen its technological footprint in Central Asia and beyond.


✔️Powered by Trade Watcher
Please open Telegram to view this post
VIEW IN TELEGRAM
22👍6🤔2
🎵 Universal and Warner near historic AI licensing deals

The music industry is on the verge of a major shift as Universal Music and Warner Music prepare to sign AI licensing agreements within weeks, according to the Financial Times.

Talks with Google, Spotify, and AI startups like Klay Vision, ElevenLabs, and Stability AI aim to set a precedent for how copyrighted songs are used to train and generate AI music. Some firms, including Suno and Udio, are already facing lawsuits from these same labels.

🖱 The proposed system would mirror streaming royalties - AI firms would pay micropayments each time a song is used or referenced.
🖱 To enable this, AI companies would need tools to track how and when music data is applied during model training and generation.
🖱 Universal and Warner’s catalogs cover global megastars like Taylor Swift, Ariana Grande, and Charli XCX, giving them strong leverage in negotiations.

If finalized, these deals could define the industry’s financial and legal framework for AI-generated music and decide who gets paid when machines start making hits.


✔️Powered by Trade Watcher
Please open Telegram to view this post
VIEW IN TELEGRAM
6👍3
📊 PitchBook found a “founder discount” that never goes away

A new PitchBook study compared startups led by first-time founders vs. repeat founders and the gap is wider than expected.

Even after controlling for growth, revenue, and market stage, companies run by first-time founders remain undervalued. The effect doesn’t fade at later rounds: even at $100M+ in revenue, startups with rookie founders still command lower valuations.

🖱 Experienced founders raise capital faster and on better terms at every stage.
🖱 Investor perception of “execution risk” persists long after traction appears.
🖱 The premium for past success acts almost like a permanent multiplier.

Makes you wonder how much higher Google’s valuation would’ve been if Page and Brin had built something before it.


✔️Powered by Trade Watcher
Please open Telegram to view this post
VIEW IN TELEGRAM
3
💰 Kraken raises $500M at $15B valuation ahead of IPO

Crypto exchange Kraken has secured $500 million in new funding at a $15 billion valuation, as it prepares for a planned IPO in 2026. The round drew multiple institutional backers, including Tribe Capital, led by Kraken co-CEO Arjun Sethi, who also invested personally.

🖱 Funding will fuel global expansion, product development, and compliance upgrades.
🖱 The round had no single lead investor, signaling broad institutional confidence.
🖱 Kraken’s valuation positions it as one of the top private exchanges globally, rivaling Coinbase pre-IPO.
🖱 Comes amid renewed institutional interest in regulated crypto platforms.
🖱 The company aims to differentiate through transparency and licensing strength.

As crypto matures, the next bull run may belong less to token traders, and more to the infrastructure players building trust.


✔️Powered by Trade Watcher
Please open Telegram to view this post
VIEW IN TELEGRAM
10
📊 Crypto Venture Weekly: September 29–October 3, 2025

$312M was raised across 18 projects this week, led by Flying Tulip’s $200M seed round and xMoney’s $21.5M strategic raise. Meanwhile, AlphaTON Capital, Strategy, and VivoPower International allocated a combined $112M into TON, BTC, and XRP treasuries.

Here’s what the top 8 are building 👇

🖱 Flying Tulip ($200M seed, $1B valuation)
Full-stack onchain exchange combining a hybrid AMM and order book model.

🖱 xMoney ($21.5M, Sui Foundation)
European payments company bridging stablecoins with global commerce.

🖱 Lava ($17.5M, Extended Series A)
Platform for bitcoin-backed loans with participation from top fintech angels.

🖱 Ethena ($14M, MEXC Ventures)
DeFi protocol behind USDe, a crypto-native stablecoin backed by delta-hedged collateral.

🖱 KGeN ($13.5M, Jump Crypto, Accel, Prosus Ventures)
Building a Verified Distribution Network across AI, DeFi, gaming, and consumer apps.

🖱 Talus ($10M, Polychain, Foresight, dao5, Inception, Sui Foundation)
Developing infrastructure for AI agents and “Prediction AI.”

🖱 Mesta ($5.5M, Circle Ventures, Paxos, Village Global)
Global fiat and stablecoin payments network connecting banks and crypto rails.

🖱 BaseVol ($6M, Altos Ventures, Neoclassic, Base Ecosystem Fund)
Options trading layer on Base using 0DTE strategies and AI agents.

Investor attention this week centered on hybrid exchanges, AI-integrated infra, and stablecoin payments, while treasury allocations and fund launches underscored rising institutional interest in onchain assets.


✔️Powered by Trade Watcher
Please open Telegram to view this post
VIEW IN TELEGRAM
12
🎬 OpenAI gives content owners control as Sora plans monetization

OpenAI is rolling out new tools for rights holders to manage how their characters and content appear in Sora, its AI video app, and introducing a revenue-sharing model for those who opt in. The update follows rapid user growth and rising copyright scrutiny.

🖱 Studios can now block or approve use of their IP within Sora-generated videos.
🖱 Rights holders who allow usage will receive a share of Sora’s revenue.
🖱 OpenAI says the monetization model will involve “trial and error” before becoming standard.
🖱 The move follows Disney and others opting out of AI content reproduction.
🖱 Sora’s short-form AI videos are already generating more content than expected.

By giving studios both control and upside, OpenAI may be crafting the first sustainable business model for AI-generated media.


✔️Powered by Trade Watcher
Please open Telegram to view this post
VIEW IN TELEGRAM
7👀3
💡 Bezos says AI is in an industrial-bubble but will benefit society

Jeff Bezos argues that AI today is riding a classic industrial bubble full of hype and overvaluation, yet its long-term impact will be transformative for society.

🖱 He sees the current frenzy as speculative excess around AI infrastructure and investment.
🖱 Despite that, Bezos believes AI will boost productivity, transform industries, and improve lives.
🖱 The innovation wave will take years, and many early bets may fail.
🖱 He warns of overinvestment in platforms without real utility.
🖱 The key winners will be those who build fundamental capabilities, not just hype.

In short: don’t mistake the bubble for the future many crashes ahead, but the underlying engine is real.


✔️Powered by Trade Watcher
Please open Telegram to view this post
VIEW IN TELEGRAM
9
🤖 Where startups actually spend their AI budgets, a16z’s top 50 list

Andreessen Horowitz analyzed real transaction data from thousands of startups to find which AI tools are getting the most money, not hype, but actual spend.

✔️Powered by Trade Watcher
Please open Telegram to view this post
VIEW IN TELEGRAM
8
⚙️ OpenAI partners with AMD to build the next generation of AI infrastructure

OpenAI has signed a multi-year deal with AMD to deploy up to 6 gigawatts of GPU compute, marking one of the largest AI hardware partnerships to date.

🖱 The first phase rolls out 1 GW of AMD Instinct MI450 GPUs starting in 2026.
🖱 AMD will issue OpenAI a warrant for up to 160 million shares, tied to hardware deployment milestones.
🖱 The partnership spans hardware, firmware, and software co-design to optimize large-scale AI systems.
🖱 It gives OpenAI a diversified GPU supply beyond NVIDIA, securing long-term compute capacity.
🖱 For AMD, it locks in a flagship AI customer and deepens its position in the AI accelerator race.

OpenAI is locking in the next decade of compute and AMD just became its second engine.


✔️Powered by Trade Watcher
Please open Telegram to view this post
VIEW IN TELEGRAM
10