Episode summary
Why 95% of AI Projects Fail, Anthropic still at $2T?, and Why AI Is Still in the First Inning
How I Invest With David Weisburd
· · 49 min · Hosts: David Weisburd · Guests: Joseph Coyne
Summary
Joseph Coyne argues that enterprise AI is still early because most companies have not yet built the data infrastructure, security, and operating habits required to make it work. He says enterprises are either scared, slow, or simply unprepared, and cites a claim that 95% of enterprise AI pilots fail because data is messy, siloed, and poorly governed. In his view, the hardest part is not model quality but turning unusable internal data into something an AI system can reliably act on.
He frames the biggest moat in AI as proprietary customer data, especially the roughly 80% of data that sits behind enterprise firewalls. That dynamic, he says, favors companies that can earn trust inside specific industries and unlock workflow data, not just generic software vendors. He points to Inspiren in senior living, Bark in child safety, and Evercoast in 4D motion capture as examples where AI creates both immediate utility and valuable new training data.
Coyne also argues that business models are likely to evolve beyond pure SaaS, toward usage-based or outcome-based pricing, but that enterprises will move slowly because budgeting habits are deeply ingrained. On valuation, he is cautious: even if Anthropic and OpenAI are huge businesses, he thinks competition, lower switching costs, and new entrants make today’s prices hard to underwrite without discipline.
Key takeaways
- Joseph Coyne says enterprise AI is still in its first inning because most companies lack the data readiness and internal muscle to deploy it well.
- He cites a claim that 95% of enterprise AI pilots fail, largely because data is siloed, messy, or not secure enough for production use.
- Coyne sees the strongest moat as proprietary data behind enterprise firewalls, not just model performance.
- He favors vertical AI businesses that combine domain expertise, trust, and workflow integration in industries like healthcare, banking, and airlines.
- He thinks AI will push pricing from pure SaaS toward hybrid usage or outcome-based models, but adoption will be gradual because enterprise buyers need budget certainty.
- Coyne prefers seed-stage investing because AI can reach product maturity faster, compressing the time between non-obvious and obvious risk.
Notable quotes
“I think specifically enterprise AI is really just starting to scratch the surface of what it can do within enterprise”
“95% of enterprise AI deployments are failing these days”
“about 80% or more of the world's data is behind firewalls”
“we always enter at seed. I think you need to create some realistic benchmarking and understanding that that will happen”
How the episode opens
So I did invest in anthropic at a $4 billion valuation. How would you underwrite it today? I — Would look at the implied forward capture of the market, right? So how big is the market likely to be and then what percentage are you expecting to capture as anthropic visav or relative to open AI and some unknown entrance. So I think that's hard to handicap on the who's coming next right conversation.
It's unknown unknown who's going to have the idea or the resources.
That's the unknown unknown. Um but you have other folks that are devoting a lot of resources to the space that have billions of dollars in reasons to be highly defensive. The Googles, the Metas, the Amazons, uh etc.
I'll tell you the bull case virtue of Lucas Swisser. He runs growth for code 2 and he talked about the three scurves of adoption. So the first scurve, the early adopters in terms of LLMs from a revenue standpoint have been developers.
What's crazy about that is that there's only 50 million developers in the world out of 8 billion people. And on top of that, only a small percentage of developers have actually implemented if you look at worldwide kind of engineers and you have all of enterprises and you have consumers. And then you could argue a fourth scurve which is agents.
So some smart people would argue that your cam is not actually 8 billion. It's 8 billion people and their agents talking to each other. So in terms of the adoption curve, we started this conversation talking about we're in the first ending of AI, it's reported that preede rounds today on average $100 million valuation.
How do you under I think so Joe you've been in around venture for nearly two decades with firms like Bane Capital Ventures, Samsung and you believe that AI is in its first ending.
From the episode’s captions: roughly the first two minutes, lightly cleaned. Keep watching on YouTube →
Topics
- enterprise ai adoption
- proprietary data moats
- vertical ai
- ai startup funding and capital efficiency
- underwriting ai competitors
Companies and tickers mentioned
- SSNLF
- Studio VC
- Bain Capital Ventures
- Samsung
- Inspiren
- Bark
- Evercoast
- Anthropic
- OpenAI
Watch the episode
Summary, takeaways and quote selection by Octagon. Quotes are short excerpts; the full conversation is in the episode.
More from How I Invest With David Weisburd
-
Hunter Walk: Why Venture is Getting Harder, Picking Great Founders, and Investing Against Consensus -
True Anomaly’s Even Rogers on Space Warfare, China’s Military Advantage, and the Defense Tech Boom -
Why Venture Is Eating Public Markets, Picking Great VCs, and the AI Boom w/Brijesh Jeevarathnam -
Federated Hermes CEO on $400 Billion in Cash, the Private Markets Boom, and AI -
Gavin Ezekowitz on AI Valuations, Asymmetric Returns, and Why Great Investors Fail -
Anthropic at $2 Trillion? The AI Bubble, and Why the Best Startups Always Look Overvalued
How I Invest With David Weisburd, this episode and its artwork belong to David Weisburd. Octagon is not affiliated with the show; this page is an independent summary.
