Wall Street bets $$$ on Claude

Plus, image AI is eating the App Store

May 06, 2026

Welcome back! The creator of the Roomba is launching a new AI-powered home robot. And this time, it's not just vacuuming your floors. Colin Angle's new startup, Familiar Machines, is building bots that can actually see, understand, and interact with your space. No more bumping into chair legs and hoping for the best. 

Will this home robot finally deliver?

— Matt, Catherine, and the Future Tools team

Could AI Change Healthcare Forever?

In the last few days, we got two new updates that are really encouraging in the world of AI, healthcare, and medicine. 

The first is from the Mayo Clinic, where researchers developed an AI model that can help specialists detect pancreatic cancer on routine abdominal CT scans up to three years before clinical diagnosis.

  • It can identify subtle signs before tumors are even visible to humans, at a stage where the disease may be easier to treat or even cure. 

  • Researchers used scans from people who eventually developed pancreatic cancer, but the scans were taken before humans could detect it.

The AI was able to find early patterns in those images. And when you’re talking about pancreatic cancer, one of the deadliest forms of cancer, the idea that AI could help detect it years earlier is super, super encouraging.

The other study came out of Harvard, showing that AI offered more accurate emergency room diagnoses than two human doctors.

  • Researchers looked at 76 patients who came into the ER and compared diagnoses from two internal medicine physicians against diagnoses generated by OpenAI’s o1 and GPT-4o models.

  • Those diagnoses were then reviewed by other physicians who didn’t know which ones came from humans and which ones came from AI. 

It turned out the AI models were more accurate more often than the humans. I don’t think this means we’re handing medical diagnosis over to AI entirely. But I do think we’re entering a world where human doctors and AI systems collaborate, almost like getting a second opinion, to help find the best possible diagnosis.

My POV: What really excites me about this is that we’re seeing AI do real-world good. 

This is the promise we’ve been hearing about for years. This is what all these companies say they’re investing billions of dollars into AI for: to solve health problems and climate issues, to help reduce poverty and hunger, and ultimately to make the world better.

So much of what we actually see is AI analyzing spreadsheets, making cartoons of Taylor Swift fighting Mickey Mouse, or solving math problems that most humans will never need solved. And that stuff is cool. I’m not saying it’s not helpful. But this is the promise we all really want: AI that makes a real impact on millions of people and families. 

Getting a productivity boost in Excel is nice, but this is the life-changing, and potentially life-saving, stuff.

— Matt

Image AI Becomes Biggest Driver of App Growth

Via TechCrunch

New data from app intelligence firm Appfigures shows that image and video model releases are driving 6.5 times more mobile app downloads than traditional chatbot upgrades—a signal that the AI app economy has moved decisively beyond text.

The numbers:

  • Google's Gemini image model (aka Nano Banana) generated more than 22 million additional downloads in 28 days, lifting the app's performance by over 4x. Despite the massive spike, it produced only $181,000 in estimated gross consumer spending.

  • ChatGPT's GPT-4o image model drove 12+ million incremental installs in the same window—roughly 4.5x higher than text-based model releases. OpenAI was the only one to convert the momentum into real revenue: approximately $70 million in gross consumer spending.

  • Meta AI's Vibes video feed added an estimated 2.6 million incremental downloads following its launch—but generated no meaningful revenue.

The gap: Aside from OpenAI, downloads and dollars are moving in opposite directions. Google drove the largest download spike but captured almost none of the commercial value. Meta brought users in but left money on the table entirely.

Why it matters: Users are clearly responding to image and video features. But the data makes clear that engagement alone isn't a business model. The real race now is who can build an image AI experience compelling enough to get people to pay for it.

Anthropic Bets on Enterprise With a $1.5B Wall Street Venture

Anthropic is making its most aggressive enterprise move yet. The company announced a new AI services firm co-founded with Blackstone, Hellman & Friedman, and Goldman Sachs—valued at $1.5 billion and structured to compete directly with the consultants and vendors that have historically owned enterprise technology adoption.

The details: 

  • The firm operates as a consulting and services arm. Anthropic engineers are embedded inside customer organizations to deeply integrate Claude into workflows.

  • Anthropic, Blackstone, and Hellman & Friedman are each putting in roughly $300 million, with Goldman Sachs adding $150 million. General Atlantic, Apollo, GIC, and Sequoia are also in. 

  • Target sectors include healthcare, financial services, manufacturing, retail, real estate, and infrastructure—all areas in which the founding partners have deep portfolio exposure and C-suite relationships.

The bigger picture: Frontier labs are no longer content to just sell API access. By embedding engineers directly into customer workflows, Anthropic is doing what McKinsey and Accenture have done for generations—owning the relationship.

2026 State of AEO Report

A year ago, most marketers weren't thinking about AI search. Now it's one of the fastest moving channels in the industry and nobody has a playbook yet.

So we built one. We surveyed hundreds of marketers to find out how they're approaching answer engine optimization, where they're investing, what's actually working, and what isn't.

The result is the 2026 State of AEO Report. Real data. Real strategies. A clear picture of where AI search is headed and how to get ahead of it.

AI image generation and editing with text prompts

Via Flux Playground

Flux Playground lets you create and modify images through simple text prompts and reference images. 

How you can use it

  • Transform written descriptions into high-quality visuals instantly

  • Edit existing images with natural language prompts

  • Maintain consistent style across a series of images

  • Iterate quickly without switching between tools

Pricing: Free

Monetize your expertise with digital knowledge products

Via Tapflow

Tapflow helps experts turn their knowledge into sellable digital products like guides and workflows. 

How you can use it

  • Create templates or courses from your existing knowledge

  • Test market demand before building a full product

  • Collaborate in real time and organize content hierarchically

  • Accept payments through multiple integrations

Pricing: Free

Jobs, announcements, and big ideas

  • OpenClaw apologizes for a botched release, planning a smaller core and LTS branch to stabilize the project.

  • Google, Microsoft, and xAI agree to let the US government review new AI models before public release.

  • Meta expands AI age assurance to automatically place suspected teens into safer Instagram and Facebook accounts.

  • Book publishers sue Meta, alleging Llama was trained on copyrighted works without permission.

  • Google adds multi-token prediction drafters to Gemma 4, accelerating inference speeds for developers.

  • NVIDIA and ServiceNow partner on autonomous AI agents for enterprise workflows.

  • OpenAI expands ChatGPT advertising with a self-serve manager and CPC bidding for marketers.

  • Anthropic releases Claude agent templates tailored for finance and insurance workflows.

Anthropic users are NOT happy, and here’s why ⤵️

That’s a wrap! See you this Friday for more.

—Matt (FutureTools.io)