Demis is rightDemis Hassabis (@demishassabis):Yann is just plain incorrect here, he’s confusing general intelligence with universal intelligence.Brains are the most exquis​ite and complex phenomena we know of in the universe (so far), and they are in fact extremely general.Obviously one can’t circumvent the no free lunch ...

Every one of us here is carrying a superco computer. It's warm. It's wet.It's squishy and we've had it ever since we were born. This supercomput runs off a cheeseburger and consumes roughly the same amount of energy as a light bulb. And no, as you're probably thinking, I'm not referring to the smartphone in your pocket. I'm referring to the brain in your head.Brains are amazing organ. Uh they allow us to understand the world we live in. We use our brains to control our bodies, to talk with one another.And w ...

Model Capabilities & Enterprise Adoption - GPT-5 represents a significant advancement in coding, writing, and healthcare, unlocking enterprise opportunities due to its improved reliability in tool calling, structured thinking, reasoning, and problem-solving [1][2] - Enterprises can adopt these models for an increasing number of use cases, with coding being a significant area of demand [2] - 92% of Fortune 500 companies were actively using chatbots shortly after the launch of ChatGPT, indicating organic adoption in the enterprise [4] Market Share & Competition - While some analyses suggest a decrease in OpenAI's enterprise market share from 50% to 25%, the company focuses on delivering the best models and products for customers [6][7] Developer Ecosystem & Enterprise Growth - OpenAI's API is actively used by over 4 million developers daily to build new products [8] - Enterprise seats grew from 3 million to 5 million in two months, showing accelerating growth and significant potential impact for both developers and enterprises [9] Infrastructure & Investment - Project Stargate is a $500 billion investment in the United States to build infrastructure for OpenAI and the country, addressing the increasing demand for AI [11] - The company is continuously investing aggressively to meet the growing demand for AI, aiming to make it more cost-approachable for enterprises and consumers [12][13] Microsoft Partnership - OpenAI values its positive relationship with Microsoft, which has been a significant infrastructure partner through Azure since before ChatGPT [17] - The company anticipates Microsoft's continued significant involvement and is actively working on defining the future of their collaboration [17][18] Talent & Mission - OpenAI attracts talent due to its mission-driven approach to building general intelligence beneficial for all of humanity [21][22]

AI Capabilities and Limitations - The industry believes Artificial General Intelligence (AGI) is not achievable in the foreseeable future [1] - AI excels in known spaces, utilizing concepts and data for tasks like discovering Newton's laws of motion or inventing the steam locomotive [9][7] - AI struggles with tasks requiring insight and intuition, such as Archimedes' principle, special relativity, or creating art like Leonardo's Mona Lisa [8] Human Intelligence - Human intelligence uniquely enables problem-solving through insight-based learning and intuition, distinct from logic-based approaches [11][14] - Insight allows humans to traverse unknown spaces, while intuition accelerates both logical and insightful processes [21] - Human beings can create value in the unknown space, which machines cannot [22] Future Implications - The industry suggests focusing on human strengths rather than competing with machines, leveraging AI to enhance value in known areas [24] - Human General Intelligence is considered significantly broader than Artificial Intelligence, even with advancements in quantum computing [23]

Company Overview - Luma AI 的使命是发展先进的多模态模型 [1] - Luma AI 拥有一支由研究人员和工程师组成的团队，致力于实现非传统的多模态 AGI 路径 [1] Leadership & Expertise - Keegan McCallum 是 Luma AI 的 ML 基础设施负责人，拥有在多家创业公司和工程领导岗位的经验 [1] - Keegan McCallum 的背景包括投资组合优化研究 [1] Event & Community Engagement - Keegan McCallum 在旧金山举行的 AI Engineer World's Fair 上分享了见解 [1] - Luma AI 通过时事通讯与社区保持联系 [1]

AI Advancement - OpenAI's general-purpose reasoning system achieved gold medal level performance on the 2025 International Math Olympiad (IMO), demonstrating significant progress in AI [1] - The achievement highlights the advancements in AI over the past decade, turning a once unrealistic dream into reality [1] - OpenAI is preparing to release GPT-5, an experimental model incorporating new research techniques for future models [2] Future Expectations - While GPT-5 is anticipated to be well-received, OpenAI does not plan to release a model with IMO gold level capability in the immediate future (many months) [2] - OpenAI's experimental reasoning LLM has achieved gold medal-level performance on the International Math Olympiad (IMO) [2]

Agentic Economy & MCP Standard - The agentic economy is emerging, where AI agents can interact, find counterparts, and purchase services from other agents, businesses, or tools [4] - MCP (Message Communication Protocol) is becoming a standard for agentic interaction, dominating the space compared to Open API and Google's A2A [8][9] - Tool discovery, a key feature of MCP, allows agents to dynamically discover and use tools based on the workflow, differentiating it from Open API [7][8] - A centralized marketplace of MCP services, like APIFY, can provide access to various services with a single API token, enabling rapid scaling of the ecosystem [12] APIFY's Role & Marketplace - APIFY is a marketplace of 5,000 tools (actors), primarily data extraction tools, with a community of creators who monetize their tools [4] - Actors are self-contained software units with defined input and output, facilitating easy integration with other systems [4][5] - APIFY has integrations with workflow automation tools and MCP, enabling AI agents to call actors from the marketplace [6][7] - APIFY enables publishing and monetization of tools or agents, providing access to a broad ecosystem of developers and visibility [23][24] Challenges & Future - Agents currently rely on human developers for access to tools and services, hindering their ability to autonomously find and purchase services [10][11] - Trust between agents and tools is a key open question, as is the overall value and reliability of autonomous tool discovery [25][26][27] - The company paid out over $4 million to creators last month, with actors generating over $500,000 per month, indicating rapid ecosystem growth [23]

Model Development & Architecture - Gemini Thinking is presented as a solution to address limitations in test-time compute, marking progress towards general intelligence [1] - The industry focuses on identifying fundamental intelligence bottlenecks within existing models and developing solutions to improve architecture or training objectives [1] Capabilities & Steerability - Recent progress in Thinking is highlighted, emphasizing both capability and steerability improvements [1] Future Directions - The document outlines the future direction of the models, indicating ongoing development and evolution [1]