The beauty of AI and agents is that it can just do things at scale and autonomously. However, it can also F-up things at scale and autonomously.
We’ve all heard about and seen enough cases that it’s clear that we’ve entered a new era, the era of guardrails, policy, and control.
Perfect example of why guardrails needed - vibe coding so cool, well until it does this 👇🏼
To his credit Amjad responded swiftly but the point remains - if coding is the most advanced example of success in AI and agentic workflows, we are going to say way more of this in the future and there needs to be umm, yeah, policies and guardrails!
And agents can also do this - hallucinate packages.
Zooming out, ServiceNow also highlighted this need for a governance layer or Control Tower as it continued to ramp sales powered by AI 👇🏻 - click through for the specific highlighted comments from the earnings call.
Not only did Bill McDermott talk about agent control towers; he also revealed that ServiceNow has 450,000 agents in production, driving over $100 million in annual savings. That’s not just a stat; it’s a signal.
It perfectly sets the stage for my conversation on the Trends with Friends podcast with investing legend Howard Lindzon and Michael Parekh. We dug into the New Rules for Founders in an AI-Native world covering moats, speed, and what it really means to build the autonomous enterprise.
At the heart of it all is one defining question the best founders are asking:
What’s the biggest company I can build with the fewest people and the most agents doing the work?
Watch here:
As always, 🙏🏼 for reading and please share with your friends and colleagues!
Scaling Startups
#more often than not it’s product but so many examples of best product doesn’t win with distribution…
#yes i do write my newsletter weekly and thoroughly enjoy it…
#speaking of liquidity, not surprising data from Peter Walker - timeline to return 1x getting longer
Enterprise Tech
#who’s actually using these tools? Read the comments - click 👇🏼 for more
#the open/closed data ecosystem will only accelerate in the next few months
#👏🏼 Astronomer - what a move to capitalize well, you know, on the recognition they never wanted - hiring Coldplay’s ex-wife to speak even better
#lots of great businesses started with automating a spreadsheet - Vanta now worth $4.15B starting with that wedge - congrats to the team on simply amazing execution
#worth repeating - the last mile in the enterprise for agents is the longest - here’s Aaron chiming in
and my comment since this is near and dear to my heart
full newsletter from April is here if you want to know what the opportunies are!
What’s 🔥 in Enterprise IT/VC #442
The Enterprise Agent Hunger Games - the agent wave 🌊 is just starting—but without the last mile of IT infra/security it’s all vibes and no production. 🧠⚙️
"By the way, if it’s not obvious, I’m super pumped about the idea and need for a whole complete infrastructure that needs to be made and reimagined as Agentic traffic grows to 10-100x human traffic."
#agree or disagree - IMO, yes for the most part, but still room for outliers…
#🤔 so 150 or so top researchers and $20B can get you your very own OpenAI!
#👀
#Github Spark is out and going after Lovable, Replit, et al!
#yeah, moving in that direction - The future of coding might look more like a terminal + AI prompt than VSCode + tabs. Vibe-driven dev flows are real — and the IDE is starting to feel like the GUI of yesterday.
#👀 first $1M day for Synthesia - Create AI videos from text in minutes. Trusted by 50.000 companies like Amazon, Accenture, Reuters.
#no one ever said that Masa was not ambitious enough - love the vision but still so early!
#playing the long game with quantum - “JP Morgan doubles down on quantum to guard against future cyber threats”
Markets
#is AI eating Google yet? Based on Sundar’s comments and results, the answer is no! Love “we continue to ship faster than ever”
#record time, under 3 months - congrats again to Protect AI!
#Who’s next?
#here we go!