In partnership with
In Today’s Issue:
💼 Anthropic teams up with Wall Street giants
🏛️ The Trump admin is exploring potential government vetting for powerful new models.
🧬 Scientists engineer living plastic that completely self-destructs
🦾 MIT students build a system that lets AI physically move your hand
✨ And more AI goodness…
Dear Readers,
Anthropic just teamed up with Blackstone, Goldman Sachs, and Hellman & Friedman to launch a $1.5 billion venture that will embed AI engineers directly inside mid-sized businesses, and if that sounds like the moment enterprise AI stops being a buzzword and starts becoming plumbing, you're reading it right.
Today's issue unpacks what this means for the companies that actually run the economy, from community banks to manufacturing floors, and why implementation capacity may now matter more than model quality. But that's just the opener. We're also looking at a White House that's suddenly reconsidering its hands-off approach to AI regulation, a team of MIT students who built a system that lets AI physically move your hand through electrical muscle stimulation, and scientists who engineered plastic that eats itself on command in six days flat.
Add in a GPU market that now devours 75% of all data center compute spending and Anthropic's rumored proactive assistant Orbit, and you've got a issue that covers everything from Wall Street boardrooms to your wrist muscles. Let's get into it.
All the best,
Kim Isenberg
🪐 AI Reveals Hidden Planetary Worlds
Astronomers using the RAVEN AI pipeline have validated 118 planets and identified more than 2,000 strong candidates in NASA TESS data, offering one of the clearest looks yet at close-orbiting exoplanets. The discoveries include ultra-short-period planets, rare worlds in the Neptunian desert, and tightly packed multi-planet systems, while also showing that roughly 9–10% of Sun-like stars host close-in planets.
What stands out is not only the scale of discovery, but how AI is shifting astronomy from simply finding planets to building cleaner population maps of where unusual worlds appear, and where they almost never do. RAVEN’s ability to detect, vet, and statistically validate signals in one workflow makes this a meaningful step toward more systematic planet hunting.
Anthropic is building Orbit, a proactive assistant for Claude Cowork that auto-generates briefings and insights from Gmail, Slack, GitHub, Calendar, Drive, and Figma, no prompting required. Users can also deploy and pin "Orbit apps" for quick access.
It's Anthropic's answer to ChatGPT Pulse, but leaning harder into dev/creative workflows with GitHub and Figma integration. Google and Perplexity are building similar proactive layers too.
Timing: Code with Claude conference is tomorrow in SF - could be the reveal. Likely Max-only at launch.
Jensen Huang argues that AI is becoming the defining industrial platform of our time, with NVIDIA positioned at the center of a new computing era where accelerated infrastructure, robotics, and AI agents reshape companies, economies, and leadership itself.
Anthropic's $1.5B Enterprise Play
Anthropic just made one of the boldest moves in enterprise AI yet. The company announced the formation of a new AI services firm alongside Blackstone, Hellman & Friedman, and Goldman Sachs, designed to bring Claude directly into the core operations of mid-sized businesses. Think community banks, regional health systems, manufacturing plants. The kind of companies that desperately need AI but don't have armies of engineers to build it themselves.
The venture, backed by roughly $1.5 billion in committed capital, will embed engineers inside companies to redesign workflows and integrate AI into core processes. Goldman and its partners plan to use their own portfolio companies as an initial proving ground before expanding to other mid-sized firms, especially in healthcare, manufacturing, financial services, and retail. Additional backers include Apollo, General Atlantic, Sequoia Capital, and GIC, making this one of the most heavily capitalized AI deployment ventures ever assembled.
The new entity shows a broader push by AI developers to expand commercial adoption beyond large enterprises. This is about building an entirely new delivery model for frontier AI.
Why it matters: This venture signals that enterprise AI is entering its next phase, moving from pilot projects at Fortune 500s to real deployments at the companies that make up the backbone of the economy. With $1.5 billion and Wall Street's deepest networks behind it, Anthropic is betting that hands-on implementation, not just better models, is the key to winning enterprise AI.
The best prompt engineers aren't typing. They're talking.
Power users figured this out early: speaking a prompt gives you 10x more context in half the time. You include the edge cases, the examples, the tone you want — because talking is fast enough that you don't skip them.
Wispr Flow captures everything you say and turns it into clean, structured text for any AI tool. Speak messy. Get polished input. Paste into ChatGPT, Claude, Cursor, or wherever you work.
89% of messages sent with zero edits. 4x faster than typing. Works system-wide on Mac, Windows, and iPhone.
The data center compute market is projected to grow from $62B in 2022 to $546B by 2029 - a 36% CAGR - with GPUs already capturing 75% of all 2025 compute spending at $174.7B, dwarfing traditional CPUs ($26.8B) and rival accelerators ($30.9B) combined. AI hasn't just reshaped the chip market - it has made the GPU the defining unit of infrastructure investment for the decade.
AI That Moves Your Body
A team of six students just won an MIT hackathon by building something that sounds like science fiction: an AI that can physically move your hand. Called Human Operator, the system is a human augmentation tool that allows AI to briefly take control of your body to help you learn and do things you normally cannot do. It combines a camera, voice input, and electrical muscle stimulation, tiny currents sent through electrodes on your wrist and fingers, to guide your hand through movements in real time.
The system processes voice input through Anthropic's Claude API, which figures out what motion is needed, then sends signals to electrodes guiding your muscles in real time. In demo videos, a hand plays a simple melody on a keyboard, waves back at someone, and curls into an OK sign, all guided by the AI. The project won the Learn Track at MIT Hard Mode 2026, a 48-hour hackathon at MIT Media Lab focused on building intelligent physical systems that sense, adapt, and respond to people.
The project draws on prior neuromuscular research at the University of Chicago's Human Computer Integration Lab, meaning this isn't just a clever hack, it stands on years of serious science. From physical rehabilitation to learning a musical instrument, the potential applications are enormous.
This project represents a fundamental shift in how humans interact with AI, moving from screens and speakers directly onto the body. If this kind of technology matures, it could redefine accessibility, physical learning, and rehabilitation in ways we're only beginning to imagine.
Claude is not just a chatbot anymore. Is your security team ready?
Claude.ai is one thing. Claude Cowork with MCP connections, running agentic workflows, taking actions across your data with ungoverned skills? That is a different conversation entirely, and most security teams are not equipped to govern it.
Harmonic Security is built to secure everything Claude offers. Full browser controls for Claude.ai, deep governance over agentic MCP workflows, and real-time visibility into what Claude is doing across your organization. So your CISO can say yes to the tools your business is already demanding.