20VC: SpaceX Launches Largest Ever IPO | OpenAI Files to Go Public | Uber Cuts 23% of HR | Lovable Hits $500M ARR | Founders Revolt Against VCs: The Fundraising Horror Stories Going Viral

Elon Musk's Fixed-Price SpaceX IPO Strategy Raises Questions

Elon Musk is taking an unconventional approach with the SpaceX IPO by setting a fixed price of $135 per share, which values the company at $1.8 trillion. This decision short-circuits the typical price discovery process where market input helps determine the opening price. Instead, Musk is dictating the price upfront, leaving the only question to be how much investors are willing to buy at that specific valuation.

This strategy introduces significant risk compared to traditional IPOs. Typically, a company's shares are priced as close as possible to the market open to account for immediate fluctuations caused by external factors like geopolitical events or other market movements. By fixing the price well in advance, there's a greater chance for error, potentially leaving money on the table if demand is overwhelming or struggling to attract orders if the set price is perceived as too high, leading to a down opening.

While it feels like an unwise move to some due to the increased uncertainty, this high-risk strategy is characteristic of Elon Musk's business philosophy. He is known for embracing challenging situations and taking on more risk. This approach aligns with his historical track record of pushing boundaries and, in this case, applying his unique vision to the public offering of SpaceX.

But the one thing we know about Elon for the last thirty years is when he hears the word more risks, he says, "Yes, please, I'll have two."

SpaceX IPO Outlook: Potential Flat Start But Strong Long-Term Growth

SpaceX's initial public offering might not experience the typical 'pop' seen in many IPOs. Unlike traditional processes where bankers aim to price shares to generate an immediate surge in value, often with an 8-10x oversubscribed book, SpaceX's fixed-price mechanism and a book that is only two times covered could lead to a flat or even slightly negative trading debut.

The initial trading day might not be dramatic, with predictions suggesting an equal chance of the stock going down, remaining flat, or going up. This is primarily due to the non-traditional pricing approach that doesn't actively account for demand in the same way a typical Wall Street-led IPO would.

Despite a potentially subdued initial trading performance, SpaceX is widely recognized as an iconic and technically impressive company of its generation. Its long-term outlook is robust, with expectations for an 'inexorable rise' driven by continuous achievements such as more satellite deployments and significant revenue growth.

The underlying technological advancements and ambitious mission position SpaceX as a valuable asset, suggesting that while day-one performance might be unremarkable, the company holds strong potential for creating generational wealth over time.

I think the IPO nominally will be a dud. I don't think it will trade up dramatically, but I do think every time there's great news, it will begin an inexorable rise up.

LP expectations rise for trillion-dollar outcomes, influencing fund sizes

Limited Partners are experiencing significant cash inflows from successful deals, which has dramatically elevated their expectations for returns. They are now routinely looking for 7-8x returns from General Partners, fundamentally shifting the benchmark for what constitutes a successful investment.

This new bar means that even 'five to eight billion dollar IPOs' are often no longer sufficient to meet LP return requirements. What were once considered 'once in a decade' trillion-dollar outcomes, such as SpaceX, now appear to be happening with greater frequency, with similar expectations for companies like Anthropic and OpenAI.

The increasing prevalence of these mega-outcomes has a direct impact on fund strategies. Larger fund sizes necessitate increasingly massive deals to generate meaningful returns for LPs, putting pressure on GPs to target and secure investments that have the potential for trillion-dollar valuations.

I don't know that little five to eight billion dollar IPOs really make it, make this math work anymore.

OpenAI Prepares for Public Offering and Advances Always-On AI Vision

OpenAI has confidentially filed for a public offering, a strategic move interpreted as a way to manage market expectations by hedging on the exact timing. This approach aims to prevent negative stories if the IPO process faces delays. The company's immense capital requirements for its large-scale operations necessitate access to public financial markets.

A precedent for expedited regulatory review is noted with SpaceX's S-1 filing, which was processed very quickly by the SEC. This rapid approval could indicate a similar swift path for OpenAI's IPO, suggesting a favorable market environment for companies seeking substantial public capital.

Beyond the financial strategy, Sam Altman is pushing OpenAI towards a vision of "persistent and always-on AI." This direction is evident in recent product developments, such as the Dreaming V3 update, which introduced a significant memory architecture upgrade. This advancement supports the long-term goal of integrating AI as a continuous and pervasive component of daily life.

The need for massive public capital underscores the ambition behind OpenAI's mission to develop advanced AI. The company is positioning itself to scale its infrastructure and research efforts, making AI a fundamental, always-present utility.

At some point you need the capital markets, the public capital markets, 'cause the scale involved is such that that's where you gotta go.

Apple's Pragmatic AI Strategy with Google Gemini Integration

Apple's decision to integrate Google Gemini into its ecosystem is not an admission of defeat in AI, but rather a pragmatic strategy to enhance the consumer experience. By leveraging Google's advanced model, Apple can focus on its core strength of controlling the handset and delivering compelling user-facing features, which ultimately drives continued device sales.

The broader industry is moving towards persistent, always-on AI experiences, a significant shift from the current browser-based interactions that are becoming increasingly dated. This evolution aims for a more seamless and integrated AI that anticipates user needs rather than requiring explicit requests.

A key aspect of this AI progression is the development of memory within models, enabling them to retain context from past interactions. This capability is crucial for providing more relevant and effective answers while also improving cost-effectiveness by reducing the need to repeatedly pass extensive context through tokens.

What matters for us, Apple, is delivering an amazing experience to our consumers, 'cause if we do that, they'll keep buying handsets.

Contrasting Consumer and Enterprise AI with Uber's HR Reductions

The opportunities for artificial intelligence differ significantly between consumer and enterprise applications. While consumers primarily seek delightful experiences and entertainment from AI in their non-working lives, businesses in the enterprise sector are driven by the pursuit of automation and efficiency. This distinction suggests a tougher market for consumer-focused AI companies like OpenAI compared to enterprise players like Anthropic.

The discussion shifts to Uber's recent decision to cut 23% of its HR department. While the company denies AI played a role, despite 95% of its engineers reportedly using AI daily, the move raises questions about the impact of AI on white-collar jobs and administrative functions. Historically, recruiting departments are often the first to see cuts during downturns.

The potential for AI to disrupt HR is significant. Some argue that an "AI VP of HR" could manage certain aspects of the job, such as evaluating performance, more effectively and impartially than a human. Such an AI could process extensive work histories and issues, potentially offering a less biased assessment than a human manager.

While specific figures like a 23% reduction due entirely to AI automation are questioned, it is acknowledged that some level of AI-driven efficiency is likely contributing to such cuts. The broader question for knowledge work is how much current tasks and jobs will be automated, with estimates ranging from 5% to 50%.

Consumers don't want to work. There's not a big market for consumers in their non-working life to do a whole bunch of complex research or using AI for productivity. They just want delightful experiences because they want to relax and entertain.

Uber Advances Robotaxi Trials in Europe as Revolut Achieves Significant Success

Uber is making strides in autonomous driving by launching new robotaxi experiments in Madrid, Europe. The company is partnering with technology providers to develop its self-driving capabilities, recognizing that driving is a prime sector for automation due to the large number of human roles involved. This renewed focus comes years after Uber initially scaled back its own autonomous vehicle project.

While autonomous driving technology is evolving slower than many expected, with leading companies like Waymo seeing gradual expansion, Uber's strategy to partner with external tech providers is crucial. This approach helps position Uber as the coordinating platform for future robotaxi fleets, mitigating the potential existential threat to its business model by integrating rather than being replaced.

Europe, sometimes perceived as a technical laggard, is actively participating in these advanced trials, exemplified by Madrid's robotaxi developments. Simultaneously, the European fintech company Revolut is lauded for its "amazing" growth and multi-billion dollar valuation, showcasing significant success in the region.

Uber's in a wonderful position to be the coordinating thing because it's the app we use, and if they add ten thousand robotaxis to the fleet, then things continue just fine.

Founders share fundraising horror stories, sparking debate on personal grudges and the nature of selling oneself

A viral Twitter thread, started by Greg Eisenberg, recently highlighted hundreds of founders sharing their negative fundraising experiences with venture capitalists. These "horror stories" included notable figures like the Cloudflare CEO, who recounted his challenging interactions with Vinod Khosla, bringing a widespread issue to light.

One perspective shared is that founders often hold deep-seated grudges from perceived slights during fundraising, a sentiment that can linger for years. However, this view also suggests that fundraising is fundamentally a sales process, akin to any other business deal where rejections and difficult interactions are common. The advice offered is to "get over it" and learn to navigate these situations with resilience.

A counter-argument emphasizes that fundraising is uniquely personal for founders. Unlike selling a product or service, founders are effectively "selling themselves" and their vision. This personal investment means that rejections and poor treatment can cut much deeper, making it a more emotionally taxing experience than typical sales encounters.

the difference is the founder in this case isn't selling their product, they're selling themselves.

The Challenges of VC Rejection and Direct Feedback

Venture capital operates with an extremely high rejection rate, typically turning down 99 out of 100 deals. This fundamental aspect makes achieving high founder satisfaction incredibly difficult, as the default experience for most founders engaging with VCs is rejection, which is inherently tough.

A notable instance of controversial feedback involved Vinod Khosla advising Cloudflare CEO Matthew Prince to consider replacing key team members, including Michelle, during a pitch. While this advice was ultimately proven incorrect by Cloudflare's subsequent success, it highlights the challenging dynamics when VCs perceive team imbalances.

The incident sparks a debate on the appropriate level and timing of directness in feedback. While some VCs might internalize concerns about a team and simply pass on a deal, others, like Khosla, are known for their bluntness. However, delivering such pointed advice during a pitch versus post-investment can significantly alter its reception and impact.

Given the sheer volume of companies VCs evaluate annually, some misjudgments and difficult interactions are inevitable. Even highly successful VCs with long careers acknowledge instances where they wish they had handled situations differently. This 'breakage' is an expected part of the high-stakes, high-volume venture capital process.

It's really hard to have high customer satisfaction when ninety-nine times out of a hundred you're going to tell the customer no.

Lean Companies Achieve Massive ARR with Small Teams

Companies like Lovable and Cursor are demonstrating a new era of extreme efficiency, achieving substantial Annual Recurring Revenue (ARR) with remarkably small workforces. Lovable recently reached $500 million in ARR with just 146 employees, while Cursor has hit $4 billion and is targeting $6 billion by year-end.

This trend challenges the traditional belief that startups inevitably expand their headcount significantly as they scale to hundreds of millions or billions in revenue. Instead, these examples highlight that it is possible to maintain a very lean and efficient operation even at massive scale.

Such efficiency creates disruption across the industry. It impacts venture investing by potentially requiring less capital for growth and also affects the job market. While it may reduce the overall number of available roles, it simultaneously drives up compensation for the highly skilled individuals within these productive, lean teams, as seen with Lovable's ability to generously compensate its staff.

We're seeing more and more examples to the contrary, and it is disruptive on many levels if you can stay as efficient as these guys are. It is disruptive to investing, it is disruptive to employees because it will shrink the number of these great roles and it will increase compensation.

AI Transforms Business Models by Substituting Labor with Intelligence Costs

AI-leveraged companies are achieving extreme efficiency by replacing significant labor costs with intelligence costs, often paid to AI providers like Anthropic or OpenAI. This allows them to operate with remarkably lean teams, sometimes as few as 146 employees, while still developing and maintaining complex products that include databases, hosting, management, and SEO functionalities.

This shift in resource allocation means that if a company dedicates 50-70% of its revenue to intelligence costs, it cannot simultaneously allocate a similar percentage to employee salaries. This creates a fundamentally different business model compared to traditional enterprises that rely heavily on large workforces for feature development and operations.

The new model generates high revenue per employee, making these AI-driven companies particularly attractive to both talent and investors. Employees benefit from being part of a small, highly leveraged team with strong economics, a stark contrast to larger, less AI-integrated companies like Salesforce with tens of thousands of employees. Investors are drawn to the high productivity and lean operational structures.

This approach allows startups to push out more features faster than previous generations, avoiding the need for extensive headcount. It challenges traditional notions of company scaling, encouraging a focus on technological leverage rather than simply adding more people, especially in the brutally competitive tech landscape.

If you're spending fifty to seventy percent of your revenue on intelligence from Anthropic or OpenAI, you don't have the option to also have fifty to seventy percent of your revenue on employees, 'cause there's just not enough room in the percentages.

The New Startup Blueprint: Half the Size for Same Revenue

The advent of AI is fundamentally reshaping the structure of new startups, allowing them to achieve significant revenue with substantially smaller workforces. Founders are increasingly opting for leaner go-to-market teams and sales organizations, moving away from the large, traditional sales forces seen in companies like Oracle or IBM.

This shift is driven by a massive increase in efficiency. While traditional enterprise companies like Salesforce might generate around $350,000 in revenue per employee and spend a small percentage on AI tokens, a new-era, intelligence-leveraging company could achieve $2.3 million per employee, even with a higher spend on AI tokens. This demonstrates a fundamentally different business model and operational efficiency.

Modern founders are intentionally building companies with extremely high revenue-per-employee targets, often aiming for $1 million to $2 million. This preference for lean, high-performing teams is a cultural choice, prioritizing quality talent and operational agility over bloat. They are willing to trade off some marginal revenue to maintain this structure.

The overall impact is a profound transformation in the startup landscape. Over the coming years, startups, including those in the B2B enterprise space, are projected to be roughly half the size they historically were while generating equivalent revenue. This represents a much larger systemic change than typical market fluctuations.

My sense is that roughly over the coming years, startups will be half the size that they used to be for revenue, including enterprise.

Elon Musk's Prescient AI Compute Strategy and Cursor Acquisition

Elon Musk demonstrated remarkable foresight by investing an estimated $20 to $30 billion into advanced AI compute infrastructure, building what was referred to as "Colossus" and "Colossus two." This massive capital expenditure was made in advance of generating revenue, driven by his strong conviction that AI would be the next significant technological trend.

This substantial, early investment strategically positioned Musk to capitalize on the surging demand for AI compute power. He found himself with extensive "gigawatts of capacity" precisely when other companies in the AI space needed it most.

Musk now sells this compute capacity to major AI competitors, including Anthropic for $950 million a month and Google for $1.25 billion a month, generating approximately $2 billion per month in revenue, totaling $24 billion annually. The acquisition of Cursor further enhances this strategy by providing a means to utilize or "backfill" the extensive server capacity.

What initially appeared to be a risky, ground-zero venture has transformed into a multi-billion dollar win, underscoring the prescience of his large-scale commitment to AI infrastructure.

He had the big picture conviction that AI mattered and he was willing to put twenty to thirty billion dollars of capital in the ground in advance of revenue because he felt this was the trend to back.

Ramp and Suno Valuations Signal a "Risk On" Market

The market is demonstrating a "risk on" sentiment, where investors are brave and willing to fund high-growth companies at significant valuations. This trend is exemplified by companies like Ramp, which recently raised $750 million at a $44 billion valuation, a tripling in value over a year, with $1 billion in ARR and positive free cash flow. While traditional banks might trade at 12 times earnings, Ramp's "AI story" and growth allow it to trade at 30 to 40 times revenue.

Suno, an AI music creator, also underscores this aggressive market. It raised $400 million at a $5.4 billion valuation, doubling its previous valuation in just six months. Despite the impressive technology, the rapid surge in valuation prompts questions about its immediate justification, even from current users who find its utility to be just barely there.

These valuation multiples for companies like Ramp and Suno highlight a significant divergence from conventional financial metrics. The market appears to be adjusting for growth, especially in the AI sector. The consistent influx of capital suggests that investors are currently operating with high confidence, pushing valuations to levels not seen in less brave market conditions.

The sustainability of this "risk on" environment is a key question. While money is abundant when investors are brave, market sentiment can shift rapidly. The current period suggests a belief that traditional rules have changed, allowing for unprecedented growth and corresponding valuations.

There'll always be money when people are brave, and there'll be nothing but treasuries when they're not.

Bending Spoons' Roll-Up Strategy for Consumer Tech Brands

Bending Spoons, an Italian company, has achieved remarkable success with a strategy of acquiring and revitalizing well-known but struggling consumer tech brands. Their portfolio includes names like Evernote, Vimeo, WeTransfer, AOL, and Eventbrite. The company has grown to 1.3 billion dollars in revenue and is reportedly filing to go public at a 20 billion dollar valuation.

The core of Bending Spoons' operating model involves drastic cost-cutting and aggressive price increases. For instance, they managed to reaccelerate Evernote's growth with only a fifth of its previous employees. Their approach is to eliminate extraneous expenditures, including significant marketing spend, and then substantially raise prices for their often 'sticky' user base.

This strategy relies heavily on extracting more value from existing, inertia-driven customers rather than focusing on organic unit growth for individual products. While their overall revenue growth is stellar, a large portion is attributed to new acquisitions. The company's playbook demonstrates how to turn around seemingly 'dead' companies by focusing on efficiency and monetization from established user bases.

Evernote was dead, and they reaccelerated the growth of Evernote with a fifth of the employees.

Bending Spoons' Strategy Leverages Acquired Consumer Apps and Price Hikes

Bending Spoons employs a distinct business model focused on acquiring established consumer internet applications like AOL, Eventbrite, Vimeo, and Evernote. Their strategy involves optimizing these platforms by reducing operational costs and then significantly increasing subscription prices for existing, sticky user bases. For example, they reportedly raised Evernote's annual pricing from $75 to $250.

This approach thrives on consumer inertia, as users of long-standing services are often reluctant to switch, even in the face of price increases. The model contrasts sharply with traditional tech companies that prioritize organic user growth; instead, Bending Spoons focuses on monetizing existing, highly loyal customer pools through efficiency and higher fees.

Despite its profitability and unique application of a private equity playbook to the consumer space, the valuation of Bending Spoons is a point of contention. The debate centers on whether a company relying heavily on inorganic growth through acquisitions and price hikes, rather than organic expansion, justifies high revenue multiples (e.g., 15-20 times revenue) typically seen in rapidly growing tech firms.

The firm has achieved significant scale, reportedly reaching $1.3 billion in revenue with 70-80% growth, although the source of this growth (organic vs. inorganic) is questioned. Their success highlights an opportunity in the consumer sector, mirroring how enterprise software companies often retain customers despite price increases, a niche many other investors overlook by focusing on pure growth metrics.

Work and finding product market fit is just a continuous mission of luck in some ways, and then the execution machine built after that requires no luck at all.

Databricks Delays IPO, Citing Manageable Capital Needs and Public Market Conditions

Databricks is choosing to remain a private company for now, despite its significant size. Typically, companies go public for three main reasons: to raise capital, to acquire other companies using their stock as currency, or to provide liquidity for shareholders. Databricks' current capital needs are considered manageable, allowing them to secure funding privately without the immediate pressure of a public offering.

Unlike capital-intensive foundation AI model companies such as OpenAI and Anthropic, which have raised tens or hundreds of billions in private rounds, Databricks operates as a software company with different financial requirements. This allows Databricks to access private capital at attractive terms, potentially achieving higher revenue multiples than some public competitors like Snowflake, and avoiding the complexities of a public listing.

The decision to stay private also considers the current public market environment. The upcoming year is expected to be "noisy" with several high-profile IPOs, including SpaceX and major AI model companies. This crowded landscape, combined with Databricks' lack of immediate "must-go-public" drivers, makes a private path more appealing for the time being.

The amount of money that you need to build a foundation model is two or three orders of magnitude more than anything else, so the imperative for those guys to go public is just different.

Microsoft pursues an independent AI strategy with new model launches

Microsoft recently launched new AI models, signifying a critical shift towards establishing its own independent control over foundational AI technology. This move indicates a recognition that relying solely on the partnership with OpenAI for such a core technology is not a sustainable long-term strategy for Microsoft.

The new models initially face limitations, such as lacking web search capabilities and not yet matching the performance of the best open-source alternatives. However, the launch marks an essential first step for Microsoft to play directly in the AI space, departing from a strategy where its AI plan was primarily to partner with OpenAI.

Microsoft's approach appears to be a long-term play, aiming to "grind their way to good enough" in AI over the next few years. This strategy mirrors the success of Azure, which, despite not always being superior to AWS, became robust enough to serve a vast corporate client base. The goal is to develop models that are sufficient for the bulk of low-end intelligence work, even if they don't surpass leaders like Anthropic or OpenAI.

This shift highlights Microsoft's determination to control its AI destiny, acknowledging that its prior reliance on OpenAI was untenable. The company is now actively investing in its own model development, accepting that direct participation is crucial, even if it means starting with models that are merely "good enough."

Can they do the Microsoft thing and grind their way to good enough over three to four years like Azure was never as good as AWS, but it was good enough for most of their corporates?

AI Oligopoly Threat and Export Controls' Unintended Impact

The AI ecosystem faces a significant risk of consolidation, potentially leading to an oligopoly where only a few foundation model players dominate. This concentration is considered the biggest threat to fostering a diverse and competitive environment.

In this context, there's an observable push to support alternatives to major players like Anthropic and OpenAI, as demonstrated by actions from companies such as Microsoft and others investing in competitive models.

Paradoxically, US export controls have had an adverse effect on the US position. These controls compelled non-US entities to innovate on their architectural designs, developing capabilities they otherwise wouldn't have prioritized.

This forced innovation, combined with their access to open-source model capabilities, has strengthened these non-US competitors. They now pose a more formidable threat than before the controls were implemented.

The single biggest threat to Nubia is consolidation of models. If we have concentration of model winning, we are in a tough space, and we want an ecosystem, not a monopoly.