In a recent video titled "How AI Is Killing SaaS Pricing," the speaker presents a compelling analysis of how artificial intelligence is fundamentally altering the economics and business models of software as a service (SaaS). The presentation references Mary Miker's AI trends report and explores the implications of AI's rapid growth on traditional software pricing strategies.
The Unprecedented Growth of AI
The speaker begins by highlighting the extraordinary growth rate of AI adoption compared to the internet. Referencing data from Mary Miker's AI trends report, they note: "This chart is showing the growth rate of ChatGPT users compared to internet users. And it's just showing that what internet took 23 years to come up to, ChatGPT has done in 3 years. It is a singularity event."
This accelerated adoption represents a fundamental shift in technology diffusion patterns. The speaker emphasizes the rapid pace of change, describing how "things are changing so fast. New models are coming on board so fast that it is almost hard to keep up with everything that is happening."
The Economics of AI: Dramatic Cost Reduction
One of the most significant developments is the dramatic reduction in AI costs. The speaker points to a staggering statistic: "You'll see that 99.7% reduction in cost over 2 years for AI inference cost."
This cost reduction is driven by various factors including open-source models and hardware improvements. The implications are profound, with the speaker asserting that "code is essentially becoming free."
The Democratization of Software Development
The speaker shares a personal anecdote that illustrates how AI is democratizing software development:
"Even in our company last year we had hired three or four engineers to build a product now. Akil is actually a CTO. I have started building applications on the weekends not just Akil and I know zero amount of programming."
This represents a significant shift in who can create software. No longer is development limited to those with formal programming skills. AI tools are enabling non-technical individuals to build applications, fundamentally altering the talent landscape in the industry.
The Deflationary Impact on Software
The speaker makes an important observation about software economics: "Software mostly deflate even before AI and software has been deflating ever since you know early 2000s and you know food inflates but software deflates."
With AI accelerating this trend, the speaker predicts that "the deflation rate of software is just going to be through the roof." This has profound implications for competitive moats and business models: "The fundamental assumptions on we're building a product that has a mode based on code and just pure software is slowly going to eradicate."
The Rise of Services Over Pure Software
In this new landscape, the speaker suggests that services will become increasingly valuable:
"There are other things that become more important in this day and age and that is relevant to pricing because if things are so cheap then you can have these competitors come in and commoditize your product far more early now than it used to be earlier."
The speaker continues: "In my consulting engagements, I've seen that services start to become more important because your services are hard to commoditize if you have a good team that maybe goes a long lot more than just the code itself."
AI vs. Traditional SaaS Business Models
The speaker highlights the fundamental differences between traditional SaaS and AI businesses:
"SaaS has historically had 80 plus% gross margins, some upfront costs. So it's a very interesting phase where SaaS had lot of high margins. This industry of SaaS has been based on annual recurring revenue. VCs give you funding for that. That is the basis of exits, right?"
In contrast, AI applications face different economic realities:
"Almost all of these foundation model companies are losing money still. They're still losing money. They're losing money in billions of dollars even though they're growing that fast. Even though the inference cost is reducing. So gross margins are still fairly low."
New Pricing Models for AI Applications
Given these differences, the speaker argues that traditional SaaS pricing models don't work for AI applications:
"AI applications are fundamentally different from SaaS. If they're so fundamentally different, they have to be charged on a usage based basis because you may start to lose money if you offer per user pricing."
This shift to usage-based pricing has cascading effects on company valuation and operations: "If you're doing usage based pricing with AI application with low gross margin, you're not going to be valued the same way as a SaaS company was valued earlier."
Recommendations for SaaS and AI Companies
The speaker concludes with clear advice for companies navigating this changing landscape:
"If you are just an AI business, you should not think of yourself like other SaaS companies thought. You should not think of your pricing like SaaS companies thought because you are not a SaaS company. You are actually a different breed."
For established SaaS companies adding AI capabilities, the speaker recommends treating AI features as a separate business unit: "If you have an existing SaaS offering… you have to treat it specially and differently."
Conclusion
The rapid advancement of AI is fundamentally altering the economics of software development and distribution. As code becomes increasingly commoditized and the barriers to software creation fall, companies must reconsider their value propositions, pricing strategies, and business models.
For SaaS executives, this presents both challenges and opportunities. The traditional high-margin, subscription-based SaaS model is being disrupted, but new possibilities are emerging in service-enhanced offerings and innovative pricing approaches. Companies that recognize the distinct economic realities of AI applications and adapt accordingly will be better positioned to thrive in this new era.