Another Crazy Day in AI: What AI Cannot Speed Up
- Mar 25
- 3 min read
Hello, AI Enthusiasts.
Here's another crazy day in AI:
Krieger on what AI changed about making products
U.S. Treasury to review AI rules in banking
Primer Group uses Gemini for everyday productivity
Some AI tools to try out
🎧 Listen to a quick breakdown of today’s stories.
TODAY'S FEATURED ITEM: Krieger on Judgment Speed and What Gets Cut
Image Credit: Wowza (created with Ideogram)
If building products is now faster than ever, what actually determines whether something works or not?
In the recent episode of Every's AI & I Podcast, Dan Shipper sits down with Mike Krieger, now working on AI product development at Anthropic. Drawing from his experience building Instagram, Krieger reflects on how product development has changed with AI tools. While it's now possible to go from idea to a fully functional product in a matter of hours, the conversation keeps returning to a more familiar challenge: deciding what to keep, what to remove, and how to shape something people will actually want to use.
Some points from the conversation that are worth sitting with:
AI speeds up the building process considerably, but it has no instinct for what to cut — and knowing what to remove is often where the real product work happens
Krieger compares fast AI-assisted builds to growing a tree indoors: it grows quickly, but never develops the strength that comes from gradual, real-world exposure
Rewrites and deletions used to be slow and costly; AI makes them far less painful, which changes how freely teams can question and redo their own work
Designing for AI agents means building software they can fully operate and extend from the ground up, not just interact with on a surface level
The strongest signal for a product worth pursuing is still someone with genuine, stubborn conviction behind it — more so than the idea itself
As AI-powered products become part of daily routines, users are expecting them to hold up consistently, not just work well under ideal conditions
AI agents with names and defined personalities tend to build a different kind of familiarity with users over time, something more personal than a standard tool
Speed has always been something builders wanted more of, and now they largely have it. But the conversation Krieger and Shipper are having suggests that speed mostly just surfaces the questions that were already there — about what users actually need, what's worth keeping, and what got added because it was easy rather than because it was right. Those questions don't get answered faster just because the tools have improved.
There's something grounding about hearing someone with Krieger's track record still working through these problems openly. It points to the fact that the technical side of building and the human side of building have always moved at different speeds. The gap between them hasn't closed. If anything, the faster the former gets, the more clearly you can see the latter.
Watch the full conversation here.
OTHER INTERESTING AI HIGHLIGHTS:
U.S. Treasury To Review AI Rules In Banking
/Ray Lewis, (News Writer), on The National News Desk, featured on KOMO News
The U.S. Treasury is opening discussions on easing regulations around AI use in banking, signaling a potential shift in how financial institutions adopt the technology. Through a series of conferences, industry leaders and regulators will explore how fewer restrictions could support innovation in areas like fraud detection and risk management. At the same time, concerns around bias, privacy, and cybersecurity remain part of the conversation. The move reflects a balancing act between accelerating AI adoption and managing its risks in the financial system.
Read more here.
Primer Group Uses Gemini For Everyday Productivity
/Google Workspace
Primer Group of Companies is turning AI into a practical, everyday tool across its operations. By integrating Google Workspace with Gemini and NotebookLM, teams are speeding up content creation, improving communication, and simplifying workflows. The company focused on small, high-impact use cases—like generating internal materials faster and creating onboarding content from existing documents. Instead of treating AI as an experiment, this approach shows how it can be embedded into daily work in a way that actually sticks.
Check it out here.
SOME AI TOOLS TO TRY OUT:
That’s a wrap on today’s Almost Daily craziness.
Catch us almost every day—almost! 😉
EXCITING NEWS:
The Another Crazy Day in AI newsletter is on LinkedIn!!!
Leveraging AI for Enhanced Content: As part of our commitment to exploring new technologies, we used AI to help curate and refine our newsletters. This enriches our content and keeps us at the forefront of digital innovation, ensuring you stay informed with the latest trends and developments.
Comments