Moonshot
Moonshot: Trading terminal or bot resource for on-chain and exchange markets.
ABAB Structured Brief
Moonshot is indexed in ABAB Crypto Map under Trading Terminals & Bots. This page keeps the official site, category, tags, and related ABAB coverage together as a searchable crypto project profile. Official domain: moonshot.com.
Related News & Analysis
William Shatner, Captain of Star Trek, Joins Moonshots Live
95-year-old William Shatner spoke to the audience at Moonshots Live hosted by Peter Diamandis, stating that the future promised by Star Trek in fiction is just around the corner for those present. When Diamandis asked if...
Moonshot AI Responds to Distillation Allegations
Moonshot AI team member Randy Xia responded that Fable was released on July 1, and K3 went live on July 15. According to the timeline, Moonshot completed model research, distillation, training, and release in ju...
White House Accuses China's Moonshot of Stealing Claude Fable Technology
...hite House Office of Science and Technology Policy, accused Moonshot AI of conducting thousands of queries on Claude Fable through a hidden platform and using distillation technology to develop the Kimi K3 model to evade...
Peter Diamandis Announces Moonshots Gathering to Award Over $5 Million to Builders and Filmmakers
Peter Diamandis announced that the Moonshots Gathering will award over $5 million to builders and filmmakers, showcasing a future worth living and actually building. The event will be held on September 25 in Los...
Palo Alto CEO Deep Dive: The Myths of Airtable's Fire Sale, Leo's Macro Hedge Fund Blowup, and the Trillion-Dollar AI Computing Arms Race
"Leo Aschenbrenner's Situational Awareness Blows Up Moonshot AI Raises $3.5B at $35B" (20VC with Harry Stebbings, featuring Nikesh Arora, CEO of Palo Alto Networks valued at $280 billion, with regular guests Rory O'Driscoll and Jason Lemkin). Here are the key points summarized: 1. Airtable acquired for $1.285 billion by Bending Spoons: valuation collapse and founder fatigue. • From $11 billion to $1.285 billion: • Airtable reached a valuation of $11 billion in 2021, ultimately selling to Italian capital firm Bending Spoons for $1.285 billion (annual revenue of about $485 million, annual growth rate of about 20%). • Market anchoring psychology: Ignoring the inflated $11 billion valuation from 2021, achieving nearly $500 million in revenue from a startup 10 years ago and exiting at over $1 billion is a remarkable business achievement. • Why traditional PE (like Thoma Bravo/Vista) did not bid: • Founder Fatigue: After layoffs, restructuring, and returning to founder mode, the founder chose to cash out in the face of the long restructuring cycle of the AI era. • Category eroded by AI: Previously, Airtable was an excellent no-code database; now developers can quickly create custom CRM/internal systems using Lovable, Cursor, or Claude Code in minutes, undermining the moat of no-code forms. • Bending Spoons' cash flow harvesting model: Skilled in acquiring mature sticky assets like Evernote, they create high cash flow machines by raising prices and cutting costs. 2. Former OpenAI researcher Leo Aschenbrenner's fund blowup: right trend, wrong portfolio. • Prodigy Leo gained fame for writing the renowned AI trend article "Situational Awareness" and raised a $225 million fund (at one point leveraging it to $4.5 billion). • Root of the blowup: Correctly identified the major trend in AI Capex (capital expenditure), but made fatal errors in portfolio construction—overlaying high leverage on extremely volatile tech assets, leading to a rapid blowup after a liquidity black swan event, with his public positions ultimately taken over by Citadel (Ken Griffin) for $16 billion. 3. Anthropic breaches three major corporate vulnerabilities and the AI cybersecurity storm. • Dimensionality reduction in AI offense and defense speed: • Anthropic's latest model autonomously discovered and breached corporate defenses in a short time. Zero-day vulnerabilities that previously took months for humans to investigate can now be found and automatically constructed into attack payloads by AI in seconds; the industry average for fixing vulnerabilities is 55 days. • Vulnerabilities are everywhere: Palo Alto Networks scanned open-source code packages over the past 14 weeks, discovering up to 14,000 unpatched vulnerabilities. • The irreplaceability of perimeter security: • Regardless of how smart the cutting-edge large models are, they are not on the defensive interception line at the network perimeter. The core of cybersecurity lies in blocking known threats at the perimeter and detecting and eliminating unknown threats through real-time AI large models (analyzing petabyte-level behavioral data) within one minute. 4. Nikesh Arora's core judgment: average intelligence is free, and corporate barriers lie in "exclusive context." • Key statement: Average intelligence will be free: • "In the long run, average intelligence will be free, and average intelligence will become increasingly smarter; only 'superior intelligence' that solves extremely complex problems like curing cancer or landing on the moon will require high fees." • Routine tasks like customer service and code completion will be rapidly popularized by low-cost open-source models (like Moonshot, open-weight models). • The lifeblood of a business: private context and data flywheel (Context is King): • General large models do not know the underlying architecture, system configuration, or reasons for the last five outages of customers. • Palo Alto Networks handles 400,000 customer cases annually, and the core task for all employees is to distill the logic and context of these human experts into a structured knowledge base (Vector DB). Once a sufficiently thick data flywheel is established, any LLM can be switched at the underlying level. 5. The trillion-dollar Capex arms race and energy/computing bottlenecks. • Extreme demand for computing and energy: • Giants like Microsoft, Amazon, Google, and Meta have seen quarterly Capex and cloud computing sales soar (adding hundreds of millions of dollars in revenue each quarter). • Land, permits, electricity, and chips have become key assets determining the winners and losers in AI over the next 3-5 years. From nuclear reactor startups (like Valor Atomics valued at $6 billion) to biogas power generation, any project capable of powering data centers is enjoying a significant valuation premium. • Palantir's explosive insights: • Palantir achieved nearly 100% explosive growth by encapsulating large models within complex business data flows. Companies are willing to pay top dollar for data flywheels that can directly provide business insights and solve real problems.
From OpenSea to OpenRouter: Alex Atala Analyzes Multi-Model Paradigms, Jevons Paradox, and Dynamic Cost Control for Enterprises
"OpenRouter CEO: Why Chinese Open Models Are Beating the US Why Enterprises Fear OpenAI & Anthropic" (20VC interview with Harry Stebbings, featuring OpenRouter co-founder and CEO Alex Atala), here are the key points summarized: 1. From OpenSea to OpenRouter: High-Concurrency Architecture and Market Evolution • Lessons from OpenSea: Alex was a co-founder of the NFT trading platform OpenSea. Early on, OpenSea experienced massive traffic surges and server downtime risks. He brought the underlying architecture experience of high concurrency, high availability (Uptime), and elastic scaling to OpenRouter, ensuring stability during model surges or service fluctuations. • Rise of Inference Providers: It was initially thought that model hosting would be monopolized by the three major cloud providers (AWS, Azure, GCP), but in reality, specialized inference providers like Fireworks and Together respond faster and perform better in deploying open-weight models (such as GLM, Kimi, DeepSeek). • Nvidia's ecosystem preference: Nvidia tends to diversify customer concentration by allocating GPU quotas to multiple inference providers, fostering a flourishing ecosystem of underlying computing power providers. 2. Multi-Model Future and AI Neurodiversity • Rejecting single-model monopoly: Advocating for "AI Neurodiversity," firmly believing that the future will not be dominated by a single model. Both enterprises and individuals need to use a combination of different models to achieve higher creativity and cost-effectiveness. • Specialization and brand intelligence: Enterprises will not rely solely on a generic model in the future but will fine-tune or train proprietary models (such as using LoRA plugins) for their core business while also utilizing other excellent open-source/closed-source models across the network. • Jevons Paradox validation: Taking GPT-5.6 / Luna as an example, after OpenAI reduced its price by 10 times, usage on the OpenRouter platform surged by 13 times. Lowering model prices does not reduce total expenditure; instead, it exponentially stimulates a larger demand for calls. 3. Why Enterprises Remain Cautious of Closed-Source Giants like OpenAI & Anthropic • Preventing vertical encroachment by giants (e.g., Claude Design vs. Figma): Model vendors have strong incentives to enter vertical application scenarios (e.g., Anthropic launching Claude Design). Enterprises worry that direct ties to closed-source giants will lead to opaque data policies, binding risks, and potential vertical replacement by the giants. • Data risks and VPC needs: Many enterprises find it difficult to fully trust closed-source vendors' data retention and privacy policies, preferring to deploy open-weight models in their own VPC (Virtual Private Cloud) or through open gateways for greater control. 4. The Competition of Open-Source Models Between China and the US: The US is Lagging • Strong momentum of Chinese open-source models: In the open-weight domain, Chinese open-source models (such as DeepSeek, GLM 5.2, Kimi/Moonshot, Qwen, etc.) have made significant breakthroughs in performance, inference efficiency, and writing capabilities. The US is currently lagging in the open-source model field. • Developer usage preferences: In the OpenRouter's ranking of open-source/open-weight model usage, Chinese open-source models have long occupied the top positions. • Distillation and catch-up strategies: Distillation is a conventional scientific method to enhance model efficiency. US Neolabs (new large model laboratories, such as Poolside, Thinking Machines) can quickly catch up through compliant distillation and reinforcement learning (RL), provided they solve the barriers to acquiring computing power. 5. Harness, Agent Architecture, and New Paradigms in Enterprise Management • Difference between Harness and Apps: Harness is built on Unix/command line principles as an Agent control layer, which is more composable, deterministic, and model-friendly than traditional API or UI-based Apps. • Orchestrator and Sub-Agent architecture: The mainstream architecture of the future will be a high-IQ "main orchestration model" coordinating the overall situation, issuing instructions to multiple low-cost, high-deterministic "open-source sub-agents" to execute standardized tasks such as classification and extraction. • Dynamic Employee Cost: Enterprise management will undergo transformation in the AI era. The inference costs incurred by employees using different models are highly dynamic, and in the future, enterprises will need to manage performance and costs based on the match between "employee output" and "AI computing power consumption costs."
A new species between universities, YC, and Hacker House: The Residency is redefining the business of "founder formation".
1. Vision and Breakthrough Point: Creating a "Hardcore Alternative" to Traditional University Education 1. Striking at the Pain Points of Traditional Higher Education • Disconnection of Traditional Universities from the Times: In an era where AI can teach all hard skills, 255 million people flock to traditional universities every year, many of whom graduate with huge debts and rush to management consulting firms like McKinsey to engage in limited "PPT jobs" that contribute little to real social innovation. • Accelerating the Global Rate of Innovation (North Star): The North Star metric of The Residency is to enhance the speed of global innovation. Its core assumption is: to gather the top 1% of the world's smartest, most ambitious, and creative young people, freeing them from university curricula, cumbersome institutional requirements, and student loan constraints, and fully supporting them in solving truly valuable hardcore real-world problems. 2. Overcoming the Most Deadly Enemy in the Startup Phase: Isolation • The Greatest Mental Drain of Entrepreneurship: What often torments founders in the early stages is not the product itself, but the extreme loneliness of making decisions alone without anyone who understands them. • High-Density Live-in Accelerator: Arranging about 22 founders in the same large geek apartment (Hacker House) for 3 to 6 months of 24/7 closed deep collaboration and communal living. 2. Evolution of Business Model and Scale Matrix 1. Reconstruction of Business Model: From "Project Membership Fees" to "Equity Binding" • Original Model: Initially, mainly charging project entry and membership fees (Program Fee), which has successfully established an excellent community and living experience over the past two years. • New Flywheel (Equity Model): Officially upgraded to a standard accelerator equity incubation model. By acquiring a small portion of early equity from startup teams, it aims to reinvest potential huge capital returns into space construction, upgrade infrastructure, and gradually achieve completely free entry for selected quality founders. 2. Global Network and Ecological Scale • 14 Global Locations: Established 14 independent Residency spaces worldwide, with the San Francisco (SF) location designated as the global flagship standard model. • $5 Billion Ecological Valuation: The total valuation of startups incubated, supported, and collaboratively accelerated within this network has exceeded $5 billion. 3. Geek Culture and Daily Operational Discipline 1. Strong Sense of Ritual and Community Cohesion • Visual Identity and Pride: Uniformly setting up iconic white neon winged signage at each House, serving as a symbol for residents to take photos and post on Twitter / LinkedIn on their first day of moving in to build their personal brand; creating large murals to record entrepreneurial highlights at stairway corners. • Hardcore Rules and Penalties for Tardiness: Everyone maintains a strict sense of time, with those late to the Monday evening meeting doing push-ups on the spot (10 for 1 minute late, 20 for 2 minutes late). 2. "Rapid Pitch and Mutual Assistance Demo" Every Monday Evening Each resident founder must take the stage in turn to complete two timed tasks: • First 60 Seconds: Rapid Elevator Pitch, refining business expression. • Last 60 Seconds: Reporting core progress from the previous week, plans for the current week, and clearly stating "specific help needed (Ask)" to all present founders. 4. Incubation Targets and Collaborative Network Achievements 1. Internal High Conversion Networking Flywheel • Amazing Conversion Effectiveness: The resident team locked in $10 million in Letters of Intent (LOI) within just 2.5 weeks through internal community and alumni referrals at The Residency. 2. Representative Records of Hardcore Projects • AI Offensive and Defensive Security Target (Fighting "Skynet-style" AI Doomsday): • Founder Background: Expelled and detained for 6 months due to hacking during university, later faced rejection in job applications and turned to entrepreneurship, successfully exited a startup in the Philippines, and then worked as a white-hat ethical hacker for government agencies in Germany. • Project Positioning: Building an autonomous AI white-hat system to prevent malicious automated hacking attacks (Cyber AI Doomsday), helping companies self-penetrate before being attacked. • Technical Breakthrough: Ranking high in various evaluation benchmarks, unlike competitors (such as Mitos) that require source code and underlying binary files for penetration, this system can achieve penetration and defense with completely zero prior information. • Space Orbital Data Center (Exnum): • Pain Point Insight: Ground data centers are facing extreme energy depletion and heat dissipation bottlenecks under the surge of AI computing power. • Orbital Solution: Deploying data centers on micro-satellites in polar sun-synchronous orbits, completely escaping Earth's shadow, enjoying continuous solar energy, and utilizing the extremely low-temperature environment of space for natural cooling. • Product Form: Creating "AWS in Orbit," running open-source intelligent agents and computing clusters directly at space nodes via API and network, directly proposing needs for AI inference computing power providers to present founders. 5. Core Mindset for Young Pioneers from Founders 1. Step Out of Physical and Psychological Islands • Actively Build a Like-Minded Circle: If you cannot physically move into a Hacker House in San Francisco or a big city, be active in online geek communities like Tech Twitter / X, and frequently engage with frontline builders to exchange ideas and provide feedback. 2. Trust Your Intuition and Dare to "Not Listen to Advice" • Break the Expectations Imposed by Others: Do not pursue what others claim "should be done" or seemingly obvious conventional choices; insist on following the field where you have a true deep conviction. • Building Belief Takes Time: A firm belief does not come overnight; it needs to be gradually solidified through continuous experimentation and practice, but as long as you take the first step, the entire exploration process is extremely worthwhile.
Harvey Co-founder Gabe Pereyra: Turned Gross Margin from -50% to Positive in One Quarter
..." based on the open-source weight model Kimi K3 released by Moonshot AI in July, and completed post-training in collaboration with the inference platform Fireworks AI. Harvey's engineering blog stated that the token pric...