MSX Pre-IPO Fund's Bet on Neuralink and Anduril: From Brain to Battlefield

Over the past seven days, the private markets have whispered a story that public crypto rarely touches: MSX's Pre-IPO Phase III fund has quietly added Neuralink and Anduril to its portfolio. The signal is not in the token price. It is in the narrative shift. An investment thesis that stretches from the human cortex to the autonomous drone swarm is not a diversification play. It is a declaration about where the next decade of computational power will be applied. And for those of us who have spent years watching the decentralization dream fragment into liquidity pools and governance theater, the question is not whether these companies will IPO. The question is whether we are building the right tools for the world they are creating. Let me start with the technical reality, stripped of the venture capital gloss. Neuralink is not a company yet in the commercial sense. It is an engineering marvel in a clinical trial costume. The N1 implant, with its 1,024 channels of flexible thread electrodes, represents a genuine leap in high-bandwidth neural capture. The R1 surgical robot, designed to weave those threads around blood vessels, solves an insertion problem that has haunted the field for two decades. As of late 2025, three patients carry the device. The FDA has granted breakthrough device designation. The CONVOY feasibility study, approved in November, aims to connect the N1 to assistive robotic arms. This is extraordinary science. It is also, from a financial engineering perspective, a revenue stream of exactly zero dollars. Anduril presents a different arithmetic. The Lattice software platform, which fuses sensor data into autonomous decision loops, has moved from prototype to production. The 2025 revenue estimate sits at one billion dollars, a 150% year-over-year increase. Contracts with the U.S. Department of Defense and the recent $1.3-1.4 billion acquisition of the 3D printing drone company Callum signal a company that is not waiting for permission. The partnership with OpenAI for counter-drone systems, announced in June, and the December award of a $250 million contract for large-scale expeditionary warfare from the Defense Innovation Unit, show a government customer with a seemingly insatiable appetite. Anduril is not selling hardware. It is selling an operating system for conflict. The core insight here, the one that keeps me awake in my Berlin apartment, is that these two companies represent the poles of AI's physical penetration. Neuralink pushes computation into the biological substrate. Anduril pushes it into the kinetic realm of warfare. Both are building the infrastructure for a world where intelligence is no longer confined to server racks. This is the real frontier. The crypto industry, with its obsessive focus on scaling transaction throughput, has been solving a problem that increasingly feels like rearranging deck chairs on a ship that is sailing toward a different ocean entirely. We are arguing about sequencer decentralization while the most consequential AI systems on the planet are being deployed by companies that do not care about your governance token. Now, let me offer the contrarian angle, because blind enthusiasm is the enemy of clear thought. The valuation math for these Pre-IPO bets is built on narrative premium, not on fundamentals. Neuralink's implied valuation of $80-100 billion, if we trust the market whispers, assigns a price-to-research ratio that would make a DeFi protocol's tokenomics look conservative. Anduril's $250-300 billion target valuation, at 14-28 times revenue, assumes the U.S. defense budget will continue to expand and that traditional prime contractors like Lockheed Martin and RTX will fail to adapt their own AI strategies. That is a bet, not a certainty. From my experience auditing whitepapers during the 2017 ICO mania, I learned that the gap between technical promise and market reality is where fortunes are lost. The same principle applies here. The technology is real. The timeline for scaled adoption is not. There is also a deeper structural risk that the Pre-IPO narrative conveniently ignores. Neuralink's invasive approach faces a credible challenge from non-invasive and minimally invasive alternatives like Synchron and Precision Neuroscience. A 16-channel stent-based device, placed via blood vessel, may lack the bandwidth of Neuralink's 1,024 channels, but it does not require drilling into the skull. If the FDA signals a preference for lower-risk interventions, Neuralink's technological lead could become a liability. Similarly, Anduril's software-defined defense model is elegant until it confronts the reality of contested electronic warfare environments. An AI that makes perfect decisions in a simulation can freeze or hallucinate when facing a sophisticated adversary jamming its sensors. The military-industrial complex is not known for tolerating unpredictability in lethal systems. I have seen this movie before. In the DeFi summer of 2020, I worked with MakerDAO developers on governance simulations, convinced that decentralized finance would reshape power structures. We built elegant models while whales accumulated MKR and captured the protocol's direction. The technology was sound. The incentive assumptions were naive. The same pattern repeats here. The technology is impressive, but the investment thesis depends on a linear progression from clinical trial to mass adoption, from defense contract to enduring franchise. Reality is rarely linear. The regulatory landscape for brain-computer interfaces is a minefield of privacy, autonomy, and identity questions. The international community has not even begun to seriously address the governance of autonomous weapons. These are not peripheral concerns. They are the core risks that could turn a 3-5x Pre-IPO return into a permanent lockup. Yet, I cannot dismiss the vision entirely. Summer fades. Builders remain. The people at Neuralink are not just chasing valuation. They are trying to give a paralyzed patient the ability to move a robotic arm with thought alone. The people at Anduril are building systems that could protect soldiers from drone swarms that no human operator could track. These are meaningful pursuits, and they deserve rigorous analysis, not cynical dismissal. The problem is that the financial infrastructure around them is repeating the same mistakes we made in crypto. It is pricing the endpoint without pricing the path. Trust no one. Verify everything. That applies to venture funds as much as it applies to smart contracts. What does this mean for the readers who have built their careers on decentralized networks? It means the next frontier is not about scaling Ethereum. It is about understanding how AI systems, both centralized and decentralized, will interact with the physical world. The protocols we build for identity, for data provenance, for secure communication, these will be the rails upon which the Neuralinks and Andurils of the future operate. The blockchain's role is not to be the center of the universe. It is to be the verification layer for a world where intelligence is everywhere. The question we should be asking is not how to get a token listed on a major exchange. The question is whether our infrastructure can handle the ethical and technical demands of a world where code reaches into the human brain and the battlefield. Gold is heavy. Code is light. But the lightest code still needs a foundation. As I watch these two companies march toward their inevitable IPOs, I am reminded that the most important work is often the least glamorous. It is the work of building the standards, the ethics, and the verification mechanisms that will govern these powerful tools. The MSX fund is placing a bet on the future of AI. I am placing a different bet. I am betting that the community of builders, the ones who understand that decentralization is a means and not an end, will rise to meet this challenge. The hardware will be built. The software will be written. But will we have the wisdom to guide it? That, more than any valuation, is the question that matters. Noise is cheap. Signal is rare. And the signal from this Pre-IPO announcement is clear: the era of pure digital abstraction is over. The physical world is the new canvas. Are we ready to paint on it?