03.28.2025

ForwardThinking 🤔

The AI Project Failure Paradox

Yet another report emerged this week, confirming that large companies continue to struggle to implement successful AI projects. This time it was Accenture, and it told the same story we heard in reports at the end of 2024. Only about a third of AI proofs of concept and experiments are making it into production.

Even worse, Accenture found that just 13% of companies had reported seeing any kind of enterprise value from these projects. Those are dismal numbers, and yet they are in line with similar research conducted by BCG, Deloitte and Futurum at the end of last year.

At this point, we know there is a huge vendor push to sell generative and agentic AI products and services. We know that there is a massive ongoing data center building program, aimed at anticipating the growing demand for resources required to run AI workloads. Yet customers seem to be lagging behind all of this activity, at least according to all this research. This is coming from four different reports from reputable firms, all finding similar results. It's hard to ignore.

Caution painted in yellow on grainy concrete wall.

Photo by Goh Rhy Yan on Unsplash

Just this week on LinkedIn, Box CEO Aaron Levie, who writes frequently about AI in the enterprise, was observing that AI had the advantage of executive backing that other technologies like cloud computing historically have lacked. “Up and down the executive chain in most companies, just a couple of years in, and it’s clear that every enterprise knows they will have an AI strategy,” Levie wrote.

The strong executive backing combined with an attempt to build coherent AI strategies makes these consistently low success rates even more baffling. Lack of executive leadership is often cited as a major factor when it comes to project failure inside large companies. Of course, it could also be attributed to some percentage of rank and file employees feeling threatened by this technology, and without full organizational buy-in projects can falter.

But perhaps what we’re seeing is actually a natural and expected lag in production. In a recent interview with economist Paul Krugman, Stanford Professor of Economics Erik Brynjolfsson suggested that the current AI production gap may be understandable and something that happens with every major technological shift — a pattern he calls the productivity J curve.

“What happens is, at first you have to reskill your workforce, put a lot of time into training, and people on their own have to figure things out. You build new business processes, even new products and services, new organizations,” he said. This should sound familiar, as many large companies (along with many of us as individuals) are experiencing the same challenges with AI today.

"What happens is, at first you have to reskill your workforce, put a lot of time into training, and people on their own have to figure things out."

As Brynjolfsson pointed out, we saw a similar dynamic at work in the 19th century with the advent of electricity. At first companies simply replaced their steam engines with electric motors, and for a while didn’t see much of a productivity gain. But over time as they figured out how to best implement the new technology, they started installing smaller motors across the factory, leading to entirely new concepts like the assembly line, which resulted in an explosion of productivity.

Today, we're still swapping old tech for new, when we truly need to reimagine how we work. Until that happens, I suspect productivity gains will remain elusive, and AI project success rates will continue to lag. Yet if history is any guide, smart managers will eventually discover their ‘assembly line moment,’ and progress should accelerate. Until then, I’m guessing we will continue to see reports with disappointing results.

~ Ron

What's new on the blog 📰

For DigiCert chief trust officer, building trust starts in-house

When I spoke to DigiCert chief trust officer, Lakshmi Hanspal, she told me that when it comes to building trust with customers, it has to start with ensuring the company's internal systems are safe and secure as possible.

“I've been able to see firsthand how trust has evolved from a compliance check box to a strategic imperative,” Hanspal told FastForward. And that warrants a person who is in charge of that.

Read the full article >>

If the Wiz deal gets past regulators, it would give Google a multicloud foothold

Wiz warranted a $32 billion price tag because it turns out that in a world where companies crave vendor flexibility, providing security across multiple clouds from a single interface fetches a mighty high price tag.

Read the full article >>

Google headquarters in Mountain View, CA

Photo by Ron Miller

With a strong data foundation, Capital One was more prepared for generative AI than most

Capital One's chief scientist and head of enterprise data and AI, Prem Natarajan has been with the company two years, but he recognizes that the work the company has done previously to be data first and in the cloud, before most companies were thinking about this, helped prepare them when generative AI and large language models came to the fore.

"Culturally, there is a deep reverence for data in the company’s DNA,” Natarajan told FastForward. “And that manifests itself in many ways such as how much data is used for decision making and how insights are organized around tools and techniques that help us get insights from data.”

Read the full article >>

Photo by Can Pac Swire on Flickr. Used under CC Attribution Share Alike 2.0 license.

ServiceNow took big Agentic AI swing with Moveworks acquisition

When ServiceNow announced its intent to acquire Moveworks earlier this month for nearly $3 billion, it put its SaaS counterparts on notice that it was dead serious about owning agentic AI in the enterprise — perhaps even going after markets like CRM where it doesn't have much presence today.

Read the full article >>

News of the Week 📣

More evidence we could be looking at a data center capacity glut

For the second time in a month, reports suggest that Microsoft could be backing off some more of its data center building commitments, both in the U.S. and Europe, a move that could suggest overcapacity.

According to Bloomberg, the company has pulled out of projects totaling 2 gigawatts of electricity capacity. That's a fairly significant retreat, and it follows reports last month that the company was pulling back on some other data center projects.

Jefferson Lab data center, showing blue servers on one side and green on the other connected by blue coaxial cables.

Photo by Jefferson Labs on Flickr. Used under Creative Commons Public Domain license.

Add to that, the fact that Microsoft CEO Satya Nadella, appearing on a podcast last month, suggested that he would be doing more leasing in 2027 and 2028, already anticipating a glut that he believes will bring down prices.

Earlier this week, Alibaba CEO Joe Tsai added fuel to that same fire, warning of a coming data center bubble. It adds up to some prominent executives anticipating that all the data center building we're seeing could lead to excess inventory, which as Nadella pointed out, could bring down prices. Definitely worth watching.

Databricks and anthropic announce partnership

As a startup like Databricks enters the final phase before a likely IPO, it needs to find ways to monetize the data sitting in its systems. AI has given the company a big boost in recent years. Consider that it raised $10 billion at the end of last year on an eye-popping $62 billion valuation.

But it takes some serious sales to live up to that kind of value, and this week the company announced a partnership with Anthropic, another startup with an almost identically large valuation.

Anthropic, best known for the Claude LLM, will work with Databricks on a five-year plan that hopes to bring in $100 million in revenue (presumably divided between the two companies). That feels like a big number, but it's just a drop in the bucket when you measure it against the combined valuations of these two organizations.

But it gives each company another sales channel and the hope is that it will drive demand with customers using both companies' products to build agents in-house. And when you throw in the fact that Anthropic has significant backing from AWS, it could lead to even more business if the chips fall right.

Coreweave IPO prices well below initial projections

CoreWeave, the cloud GPU company, officially IPOed this week, and the results could only be described as disappointing. Initially projected to price between $47 and $55 per share, the shares actually sold for $40 each, well below those earlier estimates.

If this was a bellwether for AI-related IPOs, it didn't bode well. Of course it didn't help that before the IPO there was a flurry of negative news including that the company relied heavily on Microsoft as a customer, that Microsoft was pulling back on some of its commitment, and that the founders had taken out big cash stakes from the company before going public in an unusual move. In its favor, OpenAI committed almost $12 billion in business the week before the IPO.

The latter wasn't enough for investors, however, who seem wary of the negatives and the final IPO price reflected that. The real proof, however, will be when the public markets have a chance to start trading the stock today, and over time as CoreWeave settles into life as public company.

Wall Street street sign with one way sign perched above and tall buildings in the background.

Photo by Chenyu Guan on Unsplash

Akamai wants to move AI inference to the Edge

Akamai began life early in the internet age in 1999 as a content delivery network. The job of a CDN is to move content to servers close to where the demand is going to be. The idea is to reduce the latency time it takes to get the content to end user machines. This week the company announced some new capability that does the same thing for inference.

As Jensen Huang noted at last week's Nvidia GTC conference, AI is shifting from training to inference, and Akamai thinks it can take advantage of this change, giving it a chance to compete with hyperscalers like Amazon, Microsoft and Google.

As with the CDNs of yesteryear, the Akamai solution puts "tools for platform engineers and developers to build and run AI applications and data-intensive workloads closer to end users." The product combines compute, data management, containerization and web assembly (WASM) tools to build AI applications at the edge.

What we're reading 📚

Person reading book, but all you can see is their hands holding the book. They are sitting on a hardwood floor steeped in shadow, creating an interesting look.

LivPhoto by Blaz Photo on Unsplash

Platformer News

These founders realized their bot was unsafe — and actually shut it down

Casey Newton

Pitchbook

Why a US listing might not always be the best path for Europe's startups

Leah Hodgson

Bloomberg

Wiz's $32 Billion Win Drives Unicorn M&A to Quarterly Record

Katie Roof

AI Magazine

Top 10: Women in the World of AI

Marcus Law

Look who's talking 👄

"People are talking, literally talking about $500 billion, several 100 billion dollars. I don’t think that’s entirely necessary. I think in a way, people are investing ahead of the demand that they’re seeing today, but they are projecting much bigger demand."

- Joe Tsai, Alibaba CEO on data center building boom

Please subscribe to the newsletter and feel free to share.

3350 Virginia St, Suite 262, Miami, FL 33133