The "trough of disillusionment" isn't a failure of technology—it's a failure of expectations

Generative AI is often framed as a revolution in productivity—promising the rise of the 10X developer, supercharging sales outreach, and allowing marketing and product teams to scale A/B testing like never before. But are those efficiency gains materializing yet?

We analyzed revenue per employee across companies in our AI index that provided detailed data on full-time employees. At first glance, the numbers seem to confirm the AI productivity boom: a median increase of 11% in revenue per employee from 2023 to 2024. But a deeper look suggests a more nuanced story.

Calculations using most recent available quarter (Q3-Q4 annualized revenue)

One key detail often overlooked is that revenue per employee has been rising for years—well before AI adoption accelerated. For example since its IPO, Palantir revenue per employee has grown at a 25% CAGR. In 2023, most enterprises were still in the early stages of AI adoption, assembling teams and launching initial AI-powered products. Yet, during that period, the median revenue per employee in the sampled group grew by 16%, far outpacing this year’s 10% growth.

Outliers like NVIDIA exist, but for most companies, AI's impact on productivity remains limited. This aligns with enterprise reports: outside software development, AI's role is still largely experimental. Although we are starting to see a shift from pilots to production, these deployments remain small-scale and in controlled environments.

However, AI adoption is shifting. Ten months ago, earnings calls were full of vague references—companies announcing their "first AI-powered product" or hiring a "Head of AI." Today, we're seeing more concrete results. HubSpot, for example, reported that “AI-driven support automation now handles over 35% of support tickets, with a goal to increase this to 50% in 2025.”

AI will amplify human capabilities, but adoption will take time. Early signs are promising, but it's important not to misinterpret the pace of adoption. I would argue the "trough of disillusionment" isn't a failure of technology—it's a failure of expectations.

On a positive note for founders, this reflects the reality that even the most innovative incumbents take time to adapt. This creates opportunities for startups with nimble teams and clean architectures who can reap the benefits of generative AI from the early days.

Calculations using most recent available quarter (Q3-Q4 annualized revenue)

Index Summary:

The AI Index tracks the performance of leading public companies prioritizing generative AI initiatives.

• In February, the Flybridge AI Index returned -5%.  For the prior 12 months, the Index has returned 4% to its current level of 132% since January of 2023.

• Out of the 32 active companies in the Index, 8 rose in October, while 24 declined.

•  Significant gainers include Super Micro (+25%), and Intel (+22%).  Significant decliners include C3.ai (-25%), and Astera Labs (-24%).

• The median NTM revenue multiple was 7.9x, with a median quarterly YoY revenue growth rate of 18%.

AI Insights and News:

Tech Companies Q4 2024 Earnings Highlights:

• META 

• More than 700 million monthly active users are now engaging with Meta AI

• More than 4 million advertisers now use at least one of Meta’s generative AI and creative tools, up from 1 million six months ago

• $60 billion to $65 billion CapEx in 2025

• IBM 

• IBM's Generative AI book of business reached over $5 billion inception-to-date, an increase of $2 billion quarter-over-quarter.

• Software AI book of business reached about $1 billion inception-to-date in Q4. AI-related Consulting revenue made up 4/5 of the AI book of business

• ServiceNow 

• ServiceNow saw a 150% quarter-over-quarter increase in AI-fueled Pro+ deals 

• The company is shifting to a hybrid subscription and consumption model for AI agents

• ServiceNow cited IDC forecasts that AI software spending will grow 46% in 2025, and AI platform software will hit $153 billion by 2028 at a 40.6% CAGR.

• 85% of customer self-service interactions at ServiceNow are now AI-driven

• 400,000 labor hours have been saved internally through AI-driven process automation.

• AI-driven sales prospecting has boosted conversion rates by up to 16x

• Google 

• AI-powered shopping tools contributed to a 13% increase in daily active users for Google Shopping in the U.S. in December 2024, compared to the previous year.

• Google Cloud's AI hypercomputer is being leveraged for large-scale AI workloads, with customers consuming 8x more compute capacity for AI model training and inferencing than 18 months ago.

• The company is investing $75 billion in CapEx in 2025

• AMD 

• Generated more than $5 billion in data center AI revenue in 2024

• MI400 series will focus on rack-scale AI solutions, integrating networking, CPUs, and GPUs at the silicon level

• SuperMicro 

• Liquid Cooling (DLC) is a key differentiator, expected to be adopted by over 30% of new data centers within 12 months due to AI's growing power requirements

• Hubspot

• AI-driven sales prospecting led to over 10,000 meetings for the sales team in Q4 alone.

• AI-driven support automation now handles over 35% of support tickets, with a goal to increase this to 50% in 2025.

• AI-powered sales bots resolve over 80% of website chat inquiries, improving sales efficiency.

• NICE 

• 97% of large enterprise deals (over $1M ARR) in 2024 included advanced AI solutions

• Shared that Gartner estimates that by 2028, 33% of enterprise software applications will include agentic AI, up from less than 1% in 2024

• NVIDIA 

• AI-driven revenue continues to surge: Data Center revenue reached $35.6 billion in Q4, up 93% year-over-year and 16% sequentially. For FY25, Data Center revenue was $115.2 billion, more than doubling from the prior year.

• Salesforce 

• Salesforce's AI-driven product line, Agentforce, is now a multi-billion-dollar revenue stream

• Agentforce: AI-powered automation platform for digital labor, deployed by 3,000 paying customers within 90 days of launch.

• Snowflake 

• AI and machine learning (ML) adoption continues to scale, with over 4,000 customers using Snowflake's AI and ML technology on a weekly basis

• AI-powered features contributed 3% of FY 2025 product revenue, highlighting the early traction of Snowflake’s AI offerings.

We thank Aiera, an AI-powered equity research platform, for accelerating our data and insight extraction process.

Note: These insights are focused on AI developments and priorities as discussed in management publications, earnings calls, and other company announcements.