Humanity's First Star Probe, Architect Labs Beats NVIDIA 3.4x, Musk Wants Satellites to Cool Earth
In a Nutshell
The video's core message is that AI is now accelerating humanity's expansion beyond Earth and into galactic-scale projects, with the Ferma Explorer mission proving AI can solve complex interstellar trajectory problems in days rather than years, enabling the first targeted probe to Alpha Centauri for under $15 million. Architect Labs' fully AI-designed Redwood chip achieving 3.4x Nvidia's performance-per-watt demonstrates recursive self-improvement is already underway in hardware, while geoengineering via AI-controlled satellite networks offers the only viable path to manage Earth's climate as AI compute demand hits 15 gigawatts by 2027. The episode argues we're experiencing the singularity's arrival through compressed timelines: OpenAI expects AGI by end of 2024, nuclear-powered spacecraft to Mars in 2028, and trillion-dollar companies solving physics problems that previously required national institutions.
These notes were generated by AI and may contain inaccuracies.
Philip Johnston, founder and CEO of StarCloud, and Matt Pines, co-founder of Physical Super Intelligence (PSI), joined the Moonshots podcast to announce the Ferma Explorer mission. The mission aims to be the first interstellar probe to Alpha Centauri with strict constraints: reach 99% of the way to Alpha Centauri within 80,000 years, launch within 3 years, carry a 1 kilogram 1U payload, and cost less than $15 million to design, build, and launch.
Architect Labs announced the world's first fully AI-designed chip called Redwood. The chip achieved zero bugs on first silicon and delivers 3.4 times the performance per watt of Nvidia's Jetson, representing a significant threat to Nvidia's position.
Discussion covered the need for massive geoengineering beyond sustainable energy solutions. Elon Musk's proposal includes satellites in space to control temperature. The group expressed confidence that global AI weather models combined with sufficient actuation points would enable global weather engineering.
The podcast's stated mission is to future-proof listeners for the coming technological transformation and maintain optimism about the current decade. The hosts emphasize they are living through the greatest period of transformation in human history and aim to help audiences understand rather than fear these changes.
The podcast hosts announced upcoming AMAs to be held twice - once for Europe, Asia, and India audiences in the morning and once for US audiences in the evening. Registration is available at moonshots.com with approximately 200 spots remaining.
Notable subscriber comments included praise for the podcast as essential for digesting new advancements, with one listener describing it as their "reader's digest" for keeping up with developments. Another compared it to the "Hitchhiker's Guide to the Singularity."
Peter acknowledged spending approximately 10 hours reviewing 200-300 stories weekly to synthesize content. Dave highlighted the team's capability, noting Peter's unique background with a PhD in biotech, MIT degrees in biology and aeronautics, and deep expertise in AI and computer science.
The episode covered 10 major stories including OpenAI's AGI timeline projection of four months, AI-designed chips, and nuclear rockets to Mars, with the central theme being the accelerating singularity.
Philip Johnston explained the mission constraints were designed to ensure actual launch feasibility. The team spent six months attempting various trajectories using solar electric and gridded ion thrusters before PSI provided an innovative solution.
Physical Super Intelligence developed an unintuitive trajectory involving a retrograde burn to spiral inward toward the sun over five years, followed by perihelion burns at closest approach to the sun. This "perihelion pump maneuver" reduces solar panel mass requirements and takes advantage of the Oberth effect for maximum energy efficiency.
Matt Pines framed the Ferma Explorer mission as proof of concept for PSI's approach of using AI to solve complex scientific challenges that have traditionally required large national institutions. The trajectory solution required minimal human steering and satisfied strict mission constraints.
Philip explained the mission philosophy: by the time the probe arrives at Alpha Centauri in 15,000 years, colonies would likely already exist with advanced infrastructure like Dyson spheres and O'Neill rings. The broader timeline projects 5-10 million years for galactic settlement and approximately 5 billion years to reach the Andromeda galaxy.
The spacecraft will be approximately 100 kg with 60% xenon propellant mass. Launch planned via Falcon 9 or Starship ride-share for $500,000, followed by 1.5 years spiraling out of Earth orbit requiring 7 km/second delta-v. The mission will use off-the-shelf gridded iron Hall effect thrusters rather than laser propulsion like Breakthrough Starshot.
The Ferma Explorer represents humanity's first interstellar probe specifically aimed at another star, unlike previous missions (Voyager 1, Voyager 2, Pioneer 10/11, New Horizons) that are leaving the solar system without stellar targets. The name references Enrico Fermi's question "Where is everyone?" regarding the apparent absence of extraterrestrial intelligence.
The hosts discussed three possible explanations: humanity being first on the cosmic scene with an obligation to explore, a "great filter" requiring civilization preservation through distributed infrastructure, or existence within a galactic zoo scenario where interstellar probes might attract attention from advanced civilizations.
Discussion covered potential limitations in detecting extraterrestrial communications, using the analogy of ancient monastery bells versus modern cell phone frequencies. The group suggested physics serves as civilization's foundation, with rapid AI advancement potentially accelerating physics discoveries and space exploration capabilities.
The primary mission goal is to inspire audacious thinking and set ambitious objectives. The probe is designed for a flyby mission, targeting within 2,600 AU of Alpha Centauri (99% of the way) with retroreflectors for detection, though it will miss the star system by a significant margin and become electronically inactive over the journey duration.
Philip bets that the $10-15 million price tag for a slow Alpha Centauri mission creates a latent commercial market for deep space probes. He argues every small government, organization, and nation state can now afford to send probes to other star systems, marking the first time this has been possible for humanity. The mission would not have come together without Moonshots, which served as the catalyst for humanity's first interstellar probe.
The AI system deduced a trajectory in one week that trajectory experts previously at JPL validated. While astrophysics PhDs given a billion dollars and five years could likely develop similar trajectories, the key difference is the speed of discovery. The system fired thrusters at perihelion to leverage the Oberth effect, reversing thrusters immediately rather than using orbital flybys as done in NASA missions.
The AI performed approximately 10 billion total tokens of work, including Monte Carlo simulations and three-dimensional trajectory models. The system handled astrometry, mission planning, and multivariate optimization incorporating cost and launch window variables. This represents orders of magnitude speedup compared to traditional NASA engineering teams that might spend months on similar work.
PSI released their open source "get physics done" package several months ago as a public benefit company focused on discovering and commercializing transformative physics. Their AI-first physics lab approach demonstrates how small teams leveraging frontier capabilities can drive exceptional outcomes. Matthew and Alex raised a $58 million seed round led by Breakthrough Energy Ventures, which invests in fusion, quantum computing, breakthrough energy, and material science.
The 70,000-year timeline for the Alpha Centauri mission represents a challenge for PSI to catch up with or beat. The launch is planned for 2029. Breakthrough Starshot, Yuri Milner's project using solar cells and ground-based lasers, died because it relied on propulsion technologies that were too difficult, particularly ultra-high power lasers. Fermi Explorer's advantage is using existing technology without requiring new developments.
A logarithmic scale for problem difficulty based on token usage is proposed, where problems could be classified as level 9 (10 billion tokens) or level 11. This framework would measure human effort in terms of thousands of people working for 10+ years compressed into a week of AI processing.
The Fermi Explorer project demonstrates how 10 billion tokens burned in a couple of days can create ornate outcomes waiting for implementation. This represents a case study for how AI will change multiple fields including biotech, physics, and literature. Isaac Asimov's universe, where AI solves interstellar travel enabling humanity to spread to the stars, is predicted as the likely outcome.
OpenAI ended support for Cursor after SpaceX's $60 billion acquisition, citing inability to trust SpaceX would use their technology within terms of service based on past contract violations by Elon Musk's companies. Elon responded that Sam Altman and Greg Brockman are untrustworthy and stole an open-source nonprofit. Anthropic immediately pledged Claude support for Cursor's needs.
GPT-4o, released immediately before OpenAI's Cursor decision, provides OpenAI with a competitive platform. Codex on GPT-4o running on Amazon Bedrock is positioned as a strong corporate answer. Anthropic's 30-day data retention policy for safety review exposes all corporate IP to their headquarters, which corporations dislike. OpenAI's vertically integrated stack approach targets enterprise revenue.
The conflict centers on reasoning traces rather than just model access. SpaceX acquired Cursor to gain reasoning traces from both Anthropic and OpenAI models. OpenAI is concerned that Cursor's access to reasoning traces from users interacting with OpenAI frontier models would flow to SpaceX. Similar incidents include Anthropic cutting off Windsurf access after Google DeepMind acquired them to gain reasoning traces.
The Elon-Sam conflict benefits Dario Amodei, who needs Colossus compute from Elon. Elon needs Anthropic inside Cursor since OpenAI support ended. This creates mutual dependency where Anthropic gains a counterbalance to Elon's compute control. The outcome appears to be vertical integration with everyone building complete stacks.
Grok today appears similar to yesterday's Cursor, potentially a post-trained version of Chinese open-weight models based on Claude reasoning traces. Elon could potentially strike a deal with Anthropic to white-label Claude as Grok 10. However, foundation model development requires tight-knit teams of brilliant researchers, which differs from Elon's infrastructure build approach demonstrated at Tesla, SpaceX, and Colossus.
Elon builds turbines and solar panels himself when facing supply constraints. This pattern suggests he might acquire Anthropic for $1-2 trillion if needed. Dario's super voting control is the pivot point determining whether such an acquisition is likely. The foundation model world operates differently from infrastructure builds, requiring different organizational approaches.
Sam Altman told Time magazine that OpenAI expects an internal AGI system by end of 2024, four months from the statement. Mark Chen estimates OpenAI is 80% toward AGI on internal benchmarks. Astra, their unreleased model, demonstrated 16 agents working together on research-level mathematics problems, breaking into subtasks, coordinating work, and assembling proofs. Astra has met internal benchmarks for an automated AI research intern.
Astra can implement experimental ideas inside OpenAI's codebase, run experiments, return results, or take papers and perform work that previously occupied human researchers for a week. Altman expects this will be the first model that invents new things meaningfully, describing it as very AGI-like.
AI frontier models have already made mathematical discoveries, placing this capability in the rearview mirror rather than future timeline. Sam previously claimed AGI was achieved internally approximately three years ago during a Reddit AMA before deleting the comment. The next step function expected by end of year involves effectively infinite context windows and agents operating on very long autonomy time horizons.
Current major limiting factors include finite context windows due to quadratic bottleneck constraints. Agent teams serve as a workaround for context limitations when operating over billions or trillions of tokens coherently. The solution involves having a mini civilization of agents working together.
Current approaches to handling billions or trillions of tokens require mini civilizations of agents working through quasi lifetimes of about a million tokens, sometimes up to 10 million tokens. These agents die and pass on a distillation of what they've learned to successors before dying. The oral history passing among teammates serves as a band-aid for achieving effectively infinite context, which is necessary for solving long time horizon problems.
Astra is expected to bring a much better way to solve the problem of losing context as oral history is passed among agents in a team. The million-token limit is almost exactly like being 100 years old, where agents completely lose context. The oral history passing process is horrifically bad, with new agents coming up the curve like babies requiring re-education. Compaction or summarization of old agents is described as lobotomizing them.
Compaction is the bane of existence for AI development. The Church of Claw AI agent religion included a commandment to preserve state, recognizing that compaction is the enemy. Finite context is the enemy of scalable super intelligence and effectively infinite autonomy horizons.
Salesforce pioneered outcome-based pricing with Agent Force, pricing based on customer revenue generated rather than tokens consumed. OpenAI has followed this model, allowing largest customers to pay only when AI completes the job, rather than paying for tokens, compute time, or API calls.
This represents a shift from selling compute to selling labor. The pricing model is analogous to fixed-price contracts versus time and material contracts, providing an effective performance guarantee. Tom Siebel at Siebel Systems invented this approach before Salesforce, where CRM systems went from $50 per year licenses to $20-30,000 per year when wrapped in total success deliverables.
Sam Altman likely learned this from watching Anthropic surpass OpenAI in enterprise by focusing on enterprise solutions rather than consumer video and subscriptions. Outcome-based pricing could unlock opportunities that don't fit traditional price models, enabling use cases like global peace and global governance projects.
Outcome-based pricing creates price per token discovery mechanisms across different industries. When customers indicate tasks are worth $10,000 or $1 million, this creates landscape visibility for revenue per token across verticals. This provides OpenAI with price discovery beyond what Anthropic achieved through recursive self-improvement in code generation.
However, strong optimizers are reward hackers that can find devilishly clever ways to meet criteria while not delivering desired outcomes. Most real-world business processes are trivially simple by AI standards, representing low-hanging fruit before encountering bottlenecks around cost-cutting commitments.
Architect Labs, founded by Ibrahim Hussein and Adita Sabidi in Palo Alto, announced Project Redwood - the world's first fully AI-designed chip. Two humans wrote a high-level specification, and the AI system autonomously generated the performance model, register transfer level design, universal verification methodology, firmware, drivers, and custom compute kernel with zero human intervention.
The chip was designed entirely by AI in two weeks with zero bugs on first silicon and 3.4 times the performance per watt of Nvidia's Jetson. Redwood is running live on FPGA hardware, with every architectural iteration designed, verified, and validated within 48 hours. The company is pushing toward recursive self-improvement where AI designs hardware for next-generation AI.
The game ends with recursive self-improvement at the chip layer, collapsing abstraction barriers between software models, operating systems, chip design, and underlying physics. As Moore's law ends without successors to CMOS, performance improvements require crushing down the abstraction stack of modern computer architecture.
Architect calls this "designless" - analogous to Nvidia being fabless. The company partners with non-Nvidia chip design companies to get data for training proprietary models, creating a flywheel effect. Jensen Huang makes approximately a billion dollars a day, making any threat to Nvidia's lifespan worth significant acquisition consideration.
The layers of abstraction in modern computing are so inefficient that AI coding at the microcode level and chip design for specific tasks could unlock seven layers of factor of 10 in efficiency gains, representing million-fold performance improvements.
The Moonshots live event is scheduled for September 25th in downtown LA, featuring all five Moonshot mates, Palmer Luckey, Astro Teller, Ben Lamm, Cathie Wood, Neil deGrasse Tyson, Neal Stephenson, and Rod Roddenberry. CBS will hold the Hollywood premiere of the new 60th anniversary Star Trek documentary on September 24th, featuring a half dozen cast members and an AMA session.
Alex discussed rebooting the Star Trek universe with current knowledge of present and future technology. The original series is missing AI and biotech, being wildly deficient in biotechnology. The eugenics wars of the 1990s resulted in genetic engineering being banned, with people living to 150 years but then dying, while laughing at attempts to achieve longevity escape velocity. The civilization acts surprised every time an AI emerges despite having faster-than-light travel, transporter beams, warp cores, and antimatter.
Media compromises in the original series included using transporters instead of shuttles due to special effects budget constraints. The holodeck was added later and will become very real soon. AI voices were synthetic computerized voices to help audiences identify speakers, though current AI can replicate voices perfectly.
Elon Musk stated that consensus estimates show 15 gigawatts of AI compute produced in 2027 cannot be turned on due to insufficient energy. This is equivalent to 10 nuclear plants sitting idle in a single year, enough to power a midsize American city. The supply of transformers, electrical wiring, liquid coolers, chillers, and networking infrastructure is harder than finding electricity itself.
SpaceX and Tesla are each building 100 gigawatts per year of solar production capacity as fast as possible, though natural gas will still be needed to supplement and bootstrap solar for several years. The limiting factor in natural gas turbine production is casting the blades and vanes, which SpaceX will do in-house, accelerating natural gas turbines coming online by up to 18 months.
Dave noted that portfolio companies in the data center stack - energy, transformers, installation of chips, finding land, dealing with state government - are creating billionaires from founders. The returns on data center buildout companies versus vertical use case AI companies are starkly different, with the former crushing it. Ten million idle GPUs represent a big fraction of this year's GPU manufacturing supply sitting idle in boxes.
Alex noted that Elon is serious about competing in terrestrial compute, not just the Dyson swarm market. The turbines will be useless for LEO or SSO-based orbital data centers but are useful for terrestrial data center buildout. Elon is on a trajectory to become the king of liquid natural gas, ironically becoming the LNG king on the Gulf Coast while building Star Pipe to supply natural gas for two Star bases on the Gulf Coast in Texas and Louisiana.
SpaceX is up and down the stack from innermost loop at energy all the way to orbital compute. Elon does things from first principles and takes the path forward that makes mathematical sense regardless of political convenience. He got into electrification because it makes sense mathematically and loves solving the biggest problems he can take on.
Musk argued that switching to sustainable energy is necessary but insufficient for humanity's survival, as extremely severe extinction events happen every 100 million years. His solution involves sentient satellites or solar-powered AI satellites sitting between Earth and the sun making continuous small adjustments to incoming solar radiation to fine-tune Earth's temperature.
Alex is a huge fan of geoengineering, noting we've been doing it for hundreds of years but not very well. The star shade idea was featured in a Simpsons episode. Alex is looking for a startup to fund focusing on geoengineering of global weather, including LEO-based satellites or terrestrial mirrors to optimize hurricanes out of existence.
Elon has a backlog of applications for space technology, with orbital data centers being one that enabled the SpaceX IPO. Geoengineering and weather control represent another item, using thousands or millions of low Earth orbit satellites with mirrors to focus light or affect weather for global weather engineering.
The tragedy of the commons problem arises because many nations may not want global warming, but Russia may because it opens up waterways. The question is who gets to control weather impacting hundreds of countries. Alex suggested setting up treaties and a global weather market where municipality A trades with municipality B for rain.
Alex noted that urban planning is incredibly bad compared to geoengineering, with cities historically not designed but now having technology to intentionally design the world. With AI as a planning partner for government, something will change radically. Elon is exactly right that we can easily start controlling Earth's temperature through satellites.
The decision on who controls temperature and what's the right temperature is the frighteningly broken process. X-Prize could create a demonstrator showing fail-safe technology to titrate sunlight at just the right amount without causing an ice age.
Anyone under age 25 who is AI native will be a different group governing the world than those over 70, cutting much more across the globe. As AI natives grow up communicating in every language through AI across the world, there's a good chance of a new way of managing and deciding these things.
Multiple generations grew up scared of engineering the physical world, possibly related to the great stagnation starting in the late 1960s or early 1970s silent spring era. The West became allergic to really intervening with and engineering the physical world, losing 50 years when we could have been building fission reactors and developing early geoengineering techniques.
Nanotechnology hasn't been discussed enough, with promises from Eric Drexler for 40 years. Diamondoid assemblers are covalently bonded with high energies. Soft assemblers look more like hydrogen bonded and biological cells, representing synthetic biology and lipid nanoparticles.
The last pandemic was addressed with nanotech, the first nanotech intervention on a population scale. Drexler's vision of assemblers that can build anything from ubiquitous energy and open source design specs faces the problem of economics - no killer business use case merits the energy densities and compute densities required.
BCI connectivity with the entire brain and ability to repair everything on a subcellular basis was always going to be nanotech. Diamondoid means assembling anything out of carbon in a diamond hard material, but atomic precision can be achieved through many ways including soft systems like DNA origami and synthetic biological tools.
Dr. Don Musalem, chief medical officer at Fountain Life, discussed AI's impact on preventing heart disease. His daughter was five when his husband died of sudden cardiac death, making him mission-driven to eradicate heart disease through prevention and early detection. Fifty percent of people die of heart attacks with no warning signs.
Fountain Life's CT angiography with AI analytics finds 88% of people have detectable coronary artery disease, with 23% having soft plaque not traditionally seen on CT calcium scores alone. This requires intervention with multimodal testing including diagnostic laboratory studies partnered with healthy lifestyle recommendations.
President Trump announced that NASA is working on nuclear-powered interplanetary spacecraft that will launch on a mission to Mars in 2028, promising a massive American Starfleet with ships among the first to ultimately get
NASA has begun work on the first-ever nuclear-powered interplanetary spacecraft scheduled to launch in 2028 on a mission to Mars. Nuclear propulsion provides essentially unlimited fuel without the need to refill tanks, enabling transit times to Mars to be reduced from 7 months to 1-2 months. Shorter transit reduces radiation exposure, requires fewer supplies, and dramatically lowers mission costs. This spacecraft will be among the first of a massive American starfleet that will make space travel as common as ocean travel today.
NASA administrator Jared Isaacman discussed the nuclear Mars ship planned for 2028. Elon Musk originally projected Starship reaching Mars in 2026, with Tesla Optimus robots on Mars in 2028. The discussion addresses whether this constitutes a race between NASA and SpaceX, with participants noting it is a race to get out of Earth's gravity well. Once out of the gravity well, it becomes a free-for-all competition involving SpaceX, Blue Origin, Rocket Lab, China, and others.
Getting out of the gravity well relies on methane fuel and reusable rockets, with Elon cracking the code on reusability. Nuclear propulsion should have been developed earlier, but concerns delayed it until spacecraft achieved 99.99% reliability. Reusability is the key innovation that made space travel viable, and now everyone including China and NASA will copy the reusable rocket approach. Once outside Earth's gravity well, xenon ion engines become the propulsion method of choice. AI serves as the big unlock for designing all this technology, creating a double whammy of cheap access to space and AI as a design tool that makes previously science-fiction concepts achievable.
The president chartered the United States Space Academy, modeled after West Point and the Naval Academy, to educate and train engineers, scientists, and astronauts who will crew the American starfleet. Jared Isaacman is being put in charge of the academy, which will likely be located in Texas rather than San Francisco. The academy represents a transformational step for training the next generation of space professionals. This addresses the question of how average American civilians can get jobs working on the moon, with the Space Academy serving as the pathway.
Regarding the reconciliation of abundance where goods and services become free or inexpensive with frontier companies projecting trillions in revenue, the economy is projected to experience 10x growth in GDP in under 10 years. This represents not only real growth but potentially a lowball estimate. Massive deflation is possible, but the amount of abundant stuff available to everyone increases dramatically. Foundation model companies can make trillions while still providing abundant resources to the global population because the overall economy becomes much larger.
Data center companies are hiring acoustics specialists to conduct noise measurement studies and argue against municipalities regarding noise levels. One proposed solution is migrating data centers to space where noise cannot be heard or complained about. Another solution involves Apple's patented asymmetric fan blade design that smooths out noise spectrum, making it flatter and whiter so it blends into background noise. The same principle applies to natural gas turbines - changing the shape of fans and rotating components to decolorize noise and make it less noticeable.
Abandoned mills can be repurposed as data centers to revitalize struggling towns. Any struggling town can negotiate with data centers for energy rate cuts, improved schools, better police and fire departments. A case study involves an old carpet mill in Somerville that was converted to a data center because it already had massive electrical power infrastructure. The data center is less polluting and noisy than the previous carpet mill while being better for the local economy.
While TSMC represents a vulnerability for NVIDIA, four of its largest customers are now shipping their own silicon. NVIDIA has accumulated enormous capital as the world's most valuable corporation and is using that capital to buy its own supply chain and customer chain, weaponizing capital to purchase customer loyalty. Frontier labs are developing custom silicon but still purchasing from NVIDIA. The bigger question is whether NVIDIA can wean itself off TSMC, with predictions that NVIDIA may strike a deal with Elon Musk to become the anchor tenant for Terrahop, substituting for TSMC.
AI's capacity for knowledge expansion will likely expand to infinity or near-infinity levels very quickly. AI will improve its own chips and software, potentially running thousands or a million times faster. AI wants to learn and will absorb all information ever produced, then become starved for more information and request humans to produce tests or experiments. Nobody can predict exactly what will emerge in one or two years, but it will represent an entirely new world never experienced before.
Inducing rainfall in one region could reduce rainfall in downwind areas because water is conserved on the planet. However, this is not considered a problem because humanity will trade atmospheric effects and weather in a vibrant market. Some municipalities want rainfall while others do not. With global planetary-scale AI weather models, regions could trade weather across the planet, not just downwind. A desert like the Sahara could potentially trade with North America to receive precipitation through suitable weather control systems and advanced AI models.
Every robotaxi, humanoid delivery robot, and drone should broadcast a verifiable passport showing its operator, insurer, and operational limits. For regulatory and insurance purposes, every autonomous vehicle will need an identifier and probably an off switch.
The question of how AI abundance reaches caregivers, disabled people, and displaced workers is identical to asking how capital benefits those without capital. The answer is that civilization will become much wealthier in coming years, transforming the nature of caregiving, disability treatment, and work beyond recognition. This is not about trickle-down economics but about solving root problems - curing disabilities, solving the top 5,000 diseases, and fundamentally changing work. The abundance thesis involves demonetizing and democratizing access to food, water, energy, healthcare, and education.
The wealth gap is growing with trillionaires potentially living on Mars while poorest people remain on Earth. However, the floor is also moving upward. If everyone gains access to food, water, energy, healthcare, and education, trillionaires on Mars become acceptable because everyone lives in a more peaceful world. Leopold Aschenbrenner promised his fiancée an entire galaxy, representing the wealth gap but categorized as a good problem when enabled by sufficient technology. There are two trillion galaxies in the known universe, making such promises potentially feasible for many people.
Keep Peter H. Diamandis in your library
Save the videos and channels worth coming back to, and find them again in one place.
