KevinWood

Behind the code, there is a specific kind of expertise. We are a team of researchers and engineers rooted in the labs of Tsinghua University, University of Cambridge, University of Alberta, Carnegie Mellon University, and Peking University—with experience at DeepMind, DeepSeek, xAI, and more. We've seen the limits of the current AI architectures firsthand from within the organizations that scaled them. Now, across three countries, we are building an alternative. We aren't just shipping another wrapper; we are shipping a new fundamental architecture.

We’re introducing the Cursor SDK so you can build agents with the same runtime, harness, and models that power Cursor. Run agents from CI/CD pipelines, create automations for end-to-end workflows, or embed agents directly inside your products.

This is the most OUTRAGEOUS deal I've seen in my 45 years on Wall Street. SpaceX just disclosed Musk's new compensation package: He gets up to 200 million super-voting shares if SpaceX hits a $7.5 trillion valuation, establishes a permanent human settlement of at least ONE MILLION people on Mars, and deploys roughly 100 terawatts of space-based computing power. Let me put the 100 terawatts in perspective: The entire electricity generation capacity of the United States is around 1.2 terawatts. The comp plan asks Musk to build more than 80x America's entire power grid... in orbit. This is a science fiction screenplay that somehow landed in front of the SEC. But here's why it actually matters for your portfolio... The S-1 reportedly claims a $28.5 trillion total addressable market, with over 90 percent attributed to AI. CapeFearAdvisors flagged this one cleanly: when Palantir went public, it disclosed a $119 billion TAM and the SEC reviewed and accepted it. SpaceX is claiming a market roughly 240x BIGGER. Now let's talk about what is actually being sold here: Reported 2025 revenue is approximately $15.5 billion. Starlink delivers around $11 billion of that with healthy margins, and the launch business is genuinely dominant. The problem is xAI - the AI piece doing all the heavy lifting in the trillion-dollar valuation pitch. xAI generated just $210 million of revenue in the first 3 quarters of 2025 while burning through $9.5 billion in cash. Ben Brey and Rupert Mitchell - a former Fidelity portfolio manager and a former head of equity capital markets at Goldman and Citi between them - ran a serious discounted cash flow on the actual operating businesses and arrived at roughly $400 billion. Lawrence Fossi covered their work recently and the math holds up. The IPO is being marketed at $1.75 TRILLION. The gap between what these businesses support and what Musk is asking the public to pay is roughly $1.35 trillion of pure narrative. Then layer on what we just learned last week... The New York Times investigation revealed Musk personally borrowed $500 million from SpaceX between 2018 and 2020 at rates as low as 1%, while bank prime rates sat around 5%. The same SpaceX has been used to bail out SolarCity, prop up Tesla during cash crunches, and absorb xAI when the AI losses became unmanageable. This is the same playbook he's run for two decades. Use a privately controlled entity as a personal piggy bank, and when the bills come due, find new investors to absorb the losses. The IPO is structured to keep that game going FOREVER. The Texas reincorporation strips away Delaware's fiduciary protections. Controlled-company status on the Nasdaq eliminates independent board requirements. And retail is being offered up to 30% of the offering (3x the normal allocation) because the institutions who actually do the math are quietly stepping away. Here is the part that finishes the case for me: Roughly $40 billion of the IPO proceeds are already spoken for before a single dollar reaches operations. About $23 billion retires SpaceX debt. Another $17 billion retires the high-interest debt sitting on xAI and X. This raise is not funding the future. It's just plugging existing holes that retail investors will now own. In my 45 years I've never seen a deal where the comp hurdle is colonizing another planet. I've never seen a disclosed TAM that exceeds verified comparables by two orders of magnitude. I've never seen a company asking the public to fund the retirement of debt incurred by separate private entities controlled by the same individual. Every red flag I've watched precede a major bust over four decades is sitting in this prospectus, in plain sight. The Tesla mispricing is being repeated on a far larger scale. And this time the bag is being handed directly to retail. Don't be the one holding it.

说“六个月就不需要人再工作”不是危言耸听就是断章取义，但凡用 AI 去正儿八经做个稍微复杂一点的项目就知道这这不靠谱。

A Beijing AI lab just released a model that can run 300 agents at once without any of them interfering with each other, and the mechanism they used to pull it off is something most researchers thought was years away. The lab is Moonshot AI. The founder is Yang Zhilin. The model is called Kimi K2.6, and they shipped it on April 20, 2026, under a modified MIT license. 1 trillion parameters. 32 billion active per token. Fully open weights. You can download it today. Here is the part that matters, and why almost nobody building multi-agent systems saw it coming. Every serious attempt at running multiple AI agents on the same task before this one collapsed past a certain threshold. Researchers have a name for it. Coordination hallucinations. Agents giving each other contradictory instructions. Agents stepping on each other's work. Agents reaching conclusions that looked right in isolation but made no sense when you tried to stitch them together. The more agents you added, the worse the output got. Above around 100 agents, the whole system became unstable. The standard response to this problem was to build an orchestration layer on top of a single model. An external framework that routes tasks to agents, tracks progress, and merges outputs. Every multi-agent startup in the last two years has shipped some version of this. None of them scale past a ceiling. Moonshot did something different. They trained the orchestrator as part of the model itself. The coordinator is not a wrapper. It is not a framework. It is a core capability baked into K2.6 during training. The model understands how to decompose a task into parallel subtasks. It knows how to route work to specialized sub-agents based on their skill profiles. It detects when an agent has stalled or failed, reassigns the task automatically, and merges the outputs into a single coherent result. The numbers are unlike anything that came before. 300 sub-agents running in parallel. 4,000 coordinated steps in a single autonomous run. 12 hours of continuous execution on hard engineering problems. One RL infrastructure team at Moonshot ran a K2.6-powered agent autonomously for 5 straight days managing monitoring, incident response, and system operations without human supervision. It never lost the thread. The benchmark nobody wants to talk about is BrowseComp in agent swarm mode. K2.6 scored 86.3. GPT-5.4 scored 78.4. That is not a rounding error. That is an 8-point lead on a benchmark specifically designed to test multi-agent coordination, and it was won by the only open-source model in the comparison. The counterintuitive finding is this. More agents does not automatically mean better results. Every ad-hoc multi-agent system before K2.6 proved the opposite. The value is not in the number. The value is in whether the orchestration holds together long enough and cleanly enough to turn 300 independent workers into something that behaves like a single, focused mind. The bottleneck in AI was never intelligence. It was stamina. The first model that figured out how to stay coordinated for 13 hours just made every 1-hour model obsolete.

Something strange is happening in tech. CTOs of billion dollar companies are quitting to take IC roles at Anthropic. Workday CTO -> MTS (Mar 2026) You[.]com CTO -> MTS (Mar 2026) Instagram CTO -> MTS (Jan 2026) Box CTO -> MTS (Dec 2025) Super[.]com CTO -> MTS (July 2025) Adept AI CTO -> MTS (Jan 2025) The mission is that real.

This thing automatically unplugs itself when your phone is full. No more overnight overcharging

这下是真见证历史了，阿联酋宣布单方面退出全世界最大的石油联盟 OPEC。 该组织由 12 个石油国组成，占全球 38% 原油产量，其中原油储备占全球 71.8%。过去以来，联盟成员都是一起商量，对原油增产还是减产，从而影响油价。 如今阿联酋单方面宣称退出该组织，并要增产原油，利益联盟分裂了。